Tootfinder

Opt-in global Mastodon full text search. Join the index!

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:40:54

Crosslisted article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[1/2]:
- Bridge-RAG: An Abstract Bridge Tree Based Retrieval Augmented Generation Algorithm With Cuckoo Fi...
Li, Liu, Zong, Tao, Dai, Ren, Liu, Jiang, Yang
arxiv.org/abs/2603.26668 mastoxiv.page/@arXiv_csIR_bot/
- SRAG: RAG with Structured Data Improves Vector Retrieval
Shalin Shah, Srikanth Ryali, Ramasubbu Venkatesh
arxiv.org/abs/2603.26670 mastoxiv.page/@arXiv_csIR_bot/
- LITTA: Late-Interaction and Test-Time Alignment for Visually-Grounded Multimodal Retrieval
Seonok Kim
arxiv.org/abs/2603.26683 mastoxiv.page/@arXiv_csIR_bot/
- Agentic AI for Human Resources: LLM-Driven Candidate Assessment
Yuksel, Anees, Elneima, Hewavitharana, Al-Badrashiny, Sawaf
arxiv.org/abs/2603.26710 mastoxiv.page/@arXiv_csIR_bot/
- SEAR: Schema-Based Evaluation and Routing for LLM Gateways
Zecheng Zhang, Han Zheng, Yue Xu
arxiv.org/abs/2603.26728 mastoxiv.page/@arXiv_csDB_bot/
- SleepVLM: Explainable and Rule-Grounded Sleep Staging via a Vision-Language Model
Guifeng Deng, Pan Wang, Jiquan Wang, Shuying Rao, Junyi Xie, Wanjun Guo, Tao Li, Haiteng Jiang
arxiv.org/abs/2603.26738 mastoxiv.page/@arXiv_csCV_bot/
- Aesthetic Assessment of Chinese Handwritings Based on Vision Language Models
Chen Zheng, Yuxuan Lai, Haoyang Lu, Wentao Ma, Jitao Yang, Jian Wang
arxiv.org/abs/2603.26768 mastoxiv.page/@arXiv_csCV_bot/
- Learning to Select Visual In-Context Demonstrations
Eugene Lee, Yu-Chi Lin, Jiajie Diao
arxiv.org/abs/2603.26775 mastoxiv.page/@arXiv_csLG_bot/
- CRISP: Characterizing Relative Impact of Scholarly Publications
Hannah Collison, Benjamin Van Durme, Daniel Khashabi
arxiv.org/abs/2603.26791 mastoxiv.page/@arXiv_csDL_bot/
- GroupRAG: Cognitively Inspired Group-Aware Retrieval and Reasoning via Knowledge-Driven Problem S...
Xinyi Duan, Yuanrong Tang, Jiangtao Gong
arxiv.org/abs/2603.26807 mastoxiv.page/@arXiv_csIR_bot/
- In your own words: computationally identifying interpretable themes in free-text survey data
Jenny S Wang, Aliya Saperstein, Emma Pierson
arxiv.org/abs/2603.26930 mastoxiv.page/@arXiv_csCY_bot/
- Multilingual Stutter Event Detection for English, German, and Mandarin Speech
Felix Haas, Sebastian P. Bayerl
arxiv.org/abs/2603.26939 mastoxiv.page/@arXiv_csSD_bot/
- FormalProofBench: Can Models Write Graduate Level Math Proofs That Are Formally Verified?
Ravi, Ying, Nesterov, Krishnan, Uskuplu, Xia, Aswedige, Nashold
arxiv.org/abs/2603.26996 mastoxiv.page/@arXiv_csAI_bot/
- PHONOS: PHOnetic Neutralization for Online Streaming Applications
Waris Quamer, Mu-Ruei Tseng, Ghady Nasrallah, Ricardo Gutierrez-Osuna
arxiv.org/abs/2603.27001 mastoxiv.page/@arXiv_eessAS_bo
- ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
Jovana Kondic, et al.
arxiv.org/abs/2603.27064 mastoxiv.page/@arXiv_csCV_bot/
- daVinci-LLM:Towards the Science of Pretraining
Qin, Liu, Mi, Xie, Huang, Si, Lu, Feng, Wu, Liu, Luo, Hou, Guo, Qiao, Liu
arxiv.org/abs/2603.27164 mastoxiv.page/@arXiv_csAI_bot/
- LightMover: Generative Light Movement with Color and Intensity Controls
Zhou, Wang, Kim, Shu, Yu, Hold-Geoffroy, Chaturvedi, Wu, Lin, Cohen
arxiv.org/abs/2603.27209 mastoxiv.page/@arXiv_csCV_bot/
- Self-evolving AI agents for protein discovery and directed evolution
Tan, Zhang, Li, Yu, Zhong, Zhou, Dong, Hong
arxiv.org/abs/2603.27303 mastoxiv.page/@arXiv_csAI_bot/
- Inference-Time Structural Reasoning for Compositional Vision-Language Understanding
Amartya Bhattacharya
arxiv.org/abs/2603.27349 mastoxiv.page/@arXiv_csCV_bot/
- LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications
Alexandre Cristov\~ao Maiorano
arxiv.org/abs/2603.27355 mastoxiv.page/@arXiv_csAI_bot/
- Heterogeneous Debate Engine: Identity-Grounded Cognitive Architecture for Resilient LLM-Based Eth...
Jakub Mas{\l}owski, Jaros{\l}aw A. Chudziak
arxiv.org/abs/2603.27404 mastoxiv.page/@arXiv_csAI_bot/
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 11:12:48

Replaced article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[2/5]:
- POTSA: A Cross-Lingual Speech Alignment Framework for Speech-to-Text Translation
Li, Cui, Wang, Ge, Huang, Li, Peng, Lu, Tashi, Wang, Dang
arxiv.org/abs/2511.09232 mastoxiv.page/@arXiv_csCL_bot/
- Beyond Elicitation: Provision-based Prompt Optimization for Knowledge-Intensive Tasks
Yunzhe Xu, Zhuosheng Zhang, Zhe Liu
arxiv.org/abs/2511.10465 mastoxiv.page/@arXiv_csCL_bot/
- $\pi$-Attention: Periodic Sparse Transformers for Efficient Long-Context Modeling
Dong Liu, Yanxuan Yu
arxiv.org/abs/2511.10696 mastoxiv.page/@arXiv_csCL_bot/
- Based on Data Balancing and Model Improvement for Multi-Label Sentiment Classification Performanc...
Zijin Su, Huanzhu Lyu, Yuren Niu, Yiming Liu
arxiv.org/abs/2511.14073 mastoxiv.page/@arXiv_csCL_bot/
- HEAD-QA v2: Expanding a Healthcare Benchmark for Reasoning
Alexis Correa-Guill\'en, Carlos G\'omez-Rodr\'iguez, David Vilares
arxiv.org/abs/2511.15355 mastoxiv.page/@arXiv_csCL_bot/
- Towards Hyper-Efficient RAG Systems in VecDBs: Distributed Parallel Multi-Resolution Vector Search
Dong Liu, Yanxuan Yu
arxiv.org/abs/2511.16681 mastoxiv.page/@arXiv_csCL_bot/
- Estonian WinoGrande Dataset: Comparative Analysis of LLM Performance on Human and Machine Transla...
Marii Ojastu, Hele-Andra Kuulmets, Aleksei Dorkin, Marika Borovikova, Dage S\"arg, Kairit Sirts
arxiv.org/abs/2511.17290 mastoxiv.page/@arXiv_csCL_bot/
- A Systematic Study of In-the-Wild Model Merging for Large Language Models
O\u{g}uz Ka\u{g}an Hitit, Leander Girrbach, Zeynep Akata
arxiv.org/abs/2511.21437 mastoxiv.page/@arXiv_csCL_bot/
- CREST: Universal Safety Guardrails Through Cluster-Guided Cross-Lingual Transfer
Lavish Bansal, Naman Mishra
arxiv.org/abs/2512.02711 mastoxiv.page/@arXiv_csCL_bot/
- Multilingual Medical Reasoning for Question Answering with Large Language Models
Pietro Ferrazzi, Aitor Soroa, Rodrigo Agerri
arxiv.org/abs/2512.05658 mastoxiv.page/@arXiv_csCL_bot/
- OnCoCo 1.0: A Public Dataset for Fine-Grained Message Classification in Online Counseling Convers...
Albrecht, Lehmann, Poltermann, Rudolph, Steigerwald, Stieler
arxiv.org/abs/2512.09804 mastoxiv.page/@arXiv_csCL_bot/
- Does Tone Change the Answer? Evaluating Prompt Politeness Effects on Modern LLMs: GPT, Gemini, an...
Hanyu Cai, Binqi Shen, Lier Jin, Lan Hu, Xiaojing Fan
arxiv.org/abs/2512.12812 mastoxiv.page/@arXiv_csCL_bot/
- Beg to Differ: Understanding Reasoning-Answer Misalignment Across Languages
Ovalle, Ross, Ruder, Williams, Ullrich, Ibrahim, Sagun
arxiv.org/abs/2512.22712 mastoxiv.page/@arXiv_csCL_bot/
- Activation Steering for Masked Diffusion Language Models
Adi Shnaidman, Erin Feiglin, Osher Yaari, Efrat Mentel, Amit Levi, Raz Lapid
arxiv.org/abs/2512.24143 mastoxiv.page/@arXiv_csCL_bot/
- JMedEthicBench: A Multi-Turn Conversational Benchmark for Evaluating Medical Safety in Japanese L...
Liu, Li, Niu, Zhang, Xun, Hou, Wang, Iwasawa, Matsuo, Hatakeyama-Sato
arxiv.org/abs/2601.01627 mastoxiv.page/@arXiv_csCL_bot/
- FACTUM: Mechanistic Detection of Citation Hallucination in Long-Form RAG
Dassen, Kotula, Murray, Yates, Lawrie, Kayi, Mayfield, Duh
arxiv.org/abs/2601.05866 mastoxiv.page/@arXiv_csCL_bot/
- {\dag}DAGGER: Distractor-Aware Graph Generation for Executable Reasoning in Math Problems
Zabir Al Nazi, Shubhashis Roy Dipta, Sudipta Kar
arxiv.org/abs/2601.06853 mastoxiv.page/@arXiv_csCL_bot/
- Symphonym: Universal Phonetic Embeddings for Cross-Script Name Matching
Stephen Gadd
arxiv.org/abs/2601.06932 mastoxiv.page/@arXiv_csCL_bot/
- LLMs versus the Halting Problem: Revisiting Program Termination Prediction
Sultan, Armengol-Estape, Kesseli, Vanegue, Shahaf, Adi, O'Hearn
arxiv.org/abs/2601.18987 mastoxiv.page/@arXiv_csCL_bot/
- MuVaC: A Variational Causal Framework for Multimodal Sarcasm Understanding in Dialogues
Diandian Guo, Fangfang Yuan, Cong Cao, Xixun Lin, Chuan Zhou, Hao Peng, Yanan Cao, Yanbing Liu
arxiv.org/abs/2601.20451 mastoxiv.page/@arXiv_csCL_bot/
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 11:12:28

Replaced article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[1/5]:
- Beyond In-Distribution Success: Scaling Curves of CoT Granularity for Language Model Generalization
Ru Wang, Wei Huang, Selena Song, Haoyu Zhang, Qian Niu, Yusuke Iwasawa, Yutaka Matsuo, Jiaxian Guo
arxiv.org/abs/2502.18273 mastoxiv.page/@arXiv_csCL_bot/
- Benchmarking NLP-supported Language Sample Analysis for Swiss Children's Speech
Anja Ryser, Yingqiang Gao, Sarah Ebling
arxiv.org/abs/2504.00780 mastoxiv.page/@arXiv_csCL_bot/
- Cultural Biases of Large Language Models and Humans in Historical Interpretation
Fabio Celli, Georgios Spathulas
arxiv.org/abs/2504.02572 mastoxiv.page/@arXiv_csCL_bot/
- BRIDGE: Benchmarking Large Language Models for Understanding Real-world Clinical Practice Text
Jiageng Wu, et al.
arxiv.org/abs/2504.19467 mastoxiv.page/@arXiv_csCL_bot/
- Understanding the Anchoring Effect of LLM with Synthetic Data: Existence, Mechanism, and Potentia...
Yiming Huang, Biquan Bie, Zuqiu Na, Weilin Ruan, Songxin Lei, Yutao Yue, Xinlei He
arxiv.org/abs/2505.15392 mastoxiv.page/@arXiv_csCL_bot/
- Just as Humans Need Vaccines, So Do Models: Model Immunization to Combat Falsehoods
Raza, Qureshi, Farooq, Lotif, Chadha, Pandya, Emmanouilidis
arxiv.org/abs/2505.17870 mastoxiv.page/@arXiv_csCL_bot/
- LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless Loops
Fu, Jiang, Hong, Li, Guo, Yang, Chen, Zhang
arxiv.org/abs/2506.14493 mastoxiv.page/@arXiv_csCL_bot/
- GHTM: A Graph-based Hybrid Topic Modeling Approach with a Benchmark Dataset for the Low-Resource ...
Farhana Haque, Md. Abdur Rahman, Sumon Ahmed
arxiv.org/abs/2508.00605 mastoxiv.page/@arXiv_csCL_bot/
- Link Prediction for Event Logs in the Process Industry
Anastasia Zhukova, Thomas Walton, Christian E. Lobm\"uller, Bela Gipp
arxiv.org/abs/2508.09096 mastoxiv.page/@arXiv_csCL_bot/
- AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation
Huang, Cao, Zhang, Kang, Wang, Wang, Luo, Zheng, Qian, Chen, Yu
arxiv.org/abs/2509.16952 mastoxiv.page/@arXiv_csCL_bot/
- Multi-View Attention Multiple-Instance Learning Enhanced by LLM Reasoning for Cognitive Distortio...
Jun Seo Kim, Hyemi Kim, Woo Joo Oh, Hongjin Cho, Hochul Lee, Hye Hyeon Kim
arxiv.org/abs/2509.17292 mastoxiv.page/@arXiv_csCL_bot/
- Dual-Space Smoothness for Robust and Balanced LLM Unlearning
Han Yan, Zheyuan Liu, Meng Jiang
arxiv.org/abs/2509.23362 mastoxiv.page/@arXiv_csCL_bot/
- The Rise of AfricaNLP: Contributions, Contributors, Community Impact, and Bibliometric Analysis
Tadesse Destaw Belay, et al.
arxiv.org/abs/2509.25477 mastoxiv.page/@arXiv_csCL_bot/
- Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Reco...
Srivastav, Zheng, Bezzam, Le Bihan, Koluguri, \.Zelasko, Majumdar, Moumen, Gandhi
arxiv.org/abs/2510.06961 mastoxiv.page/@arXiv_csCL_bot/
- Neuron-Level Analysis of Cultural Understanding in Large Language Models
Taisei Yamamoto, Ryoma Kumon, Danushka Bollegala, Hitomi Yanaka
arxiv.org/abs/2510.08284 mastoxiv.page/@arXiv_csCL_bot/
- CLMN: Concept based Language Models via Neural Symbolic Reasoning
Yibo Yang
arxiv.org/abs/2510.10063 mastoxiv.page/@arXiv_csCL_bot/
- Schema for In-Context Learning
Chen, Chen, Wang, Leong, Fung, Bernales, Aspuru-Guzik
arxiv.org/abs/2510.13905 mastoxiv.page/@arXiv_csCL_bot/
- Evaluating Latent Knowledge of Public Tabular Datasets in Large Language Models
Matteo Silvestri, Fabiano Veglianti, Flavio Giorgi, Fabrizio Silvestri, Gabriele Tolomei
arxiv.org/abs/2510.20351 mastoxiv.page/@arXiv_csCL_bot/
- LuxIT: A Luxembourgish Instruction Tuning Dataset from Monolingual Seed Data
Julian Valline, Cedric Lothritz, Siwen Guo, Jordi Cabot
arxiv.org/abs/2510.24434 mastoxiv.page/@arXiv_csCL_bot/
- Surfacing Subtle Stereotypes: A Multilingual, Debate-Oriented Evaluation of Modern LLMs
Muhammed Saeed, Muhammad Abdul-mageed, Shady Shehata
arxiv.org/abs/2511.01187 mastoxiv.page/@arXiv_csCL_bot/
toXiv_bot_toot

@arXiv_econTH_bot@mastoxiv.page
2026-03-31 11:08:56

Replaced article(s) found for econ.TH. arxiv.org/list/econ.TH/new
[1/1]:
- Reputational cheap talk: influentialness and welfare
Allen Vong
arxiv.org/abs/2505.11877 mastoxiv.page/@arXiv_econTH_bo
- Local Strategy-proofness and Dictatorship
Abinash Panda, Anup Pramanik, Ragini Saxena
arxiv.org/abs/2507.00913 mastoxiv.page/@arXiv_econTH_bo
- Endogenous Inequality Aversion: Decision criteria for triage and other ethical tradeoffs
Federico Echenique, Teddy Mekonnen, M. Bumin Yenmez
arxiv.org/abs/2601.22250 mastoxiv.page/@arXiv_econTH_bo
- Generalized Multidimensional Contests with Asymmetric Players: Equilibrium and Optimal Prize Design
Siyuan Fan, Zhonghong Kuang, Jingfeng Lu
arxiv.org/abs/2602.21564 mastoxiv.page/@arXiv_econTH_bo
- Stable Matchings with Choice Correspondences Under Acyclicity
Varun Bansal, Mihir Bhattacharya, Ojasvi Khare
arxiv.org/abs/2603.23038 mastoxiv.page/@arXiv_econTH_bo
- Calibrated Forecasting and Persuasion
Atulya Jain, Vianney Perchet
arxiv.org/abs/2406.15680 mastoxiv.page/@arXiv_csGT_bot/
- Feedback-Coupled Memory Systems: A Dynamical Model for Adaptive Coordination
Stefano Grassi
arxiv.org/abs/2603.11560 mastoxiv.page/@arXiv_csMA_bot/
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 09:33:29

Replaced article(s) found for math.LO. arxiv.org/list/math.LO/new
[1/1]:
- $\aleph_1$-free abelian non-Archimedean Polish groups
Gianluca Paolini, Saharon Shelah
arxiv.org/abs/2410.02485 mastoxiv.page/@arXiv_mathLO_bo
- Procountable groups are not classifiable by countable structures
Su Gao, Andr\'e Nies, Gianluca Paolini
arxiv.org/abs/2512.12256 mastoxiv.page/@arXiv_mathLO_bo
- Janus-faces of temporal constraint languages: a dichotomy of expressivity
Johanna Brunar, Michael Pinsker, Moritz Sch\"obi
arxiv.org/abs/2509.04347 mastoxiv.page/@arXiv_csLO_bot/
- On the word problem for just infinite groups
Alexey Talambutsa
arxiv.org/abs/2512.24266 mastoxiv.page/@arXiv_mathGR_bo
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-31 10:46:18

Crosslisted article(s) found for math.CT. arxiv.org/list/math.CT/new
[1/1]:
- Contraherent cosheaves of contramodules on Noetherian formal schemes
Leonid Positselski
arxiv.org/abs/2603.27732 mastoxiv.page/@arXiv_mathAG_bo
- Ribbon categories from ind-exact algebras: simple current case
Kenichi Shimizu, Harshit Yadav
arxiv.org/abs/2603.28215 mastoxiv.page/@arXiv_mathQA_bo
- Anick Resolution for Lawvere Theories from Algebraic Discrete Morse Theory
Mirai Ikebuchi
arxiv.org/abs/2603.28382 mastoxiv.page/@arXiv_mathKT_bo
- Categorical Time-Reversal Symmetries
Rui Wen, Sakura Schafer-Nameki
arxiv.org/abs/2603.28720 mastoxiv.page/@arXiv_condmatst
toXiv_bot_toot

@arXiv_mathSG_bot@mastoxiv.page
2026-03-31 10:49:48

Crosslisted article(s) found for math.SG. arxiv.org/list/math.SG/new
[1/1]:
- Weak supermajorization between symplectic spectra of positive definite matrix and its pinching
Temjensangba, Hemant Kumar Mishra
arxiv.org/abs/2603.27634 mastoxiv.page/@arXiv_mathFA_bo
- Log-concavity from enumerative geometry of planar curve singularities
Tao Su, Baiting Xie, Chenglong Yu
arxiv.org/abs/2603.27888 mastoxiv.page/@arXiv_mathAG_bo
- The $\chi_y$-genus, Chern number inequalities and signature
Ping Li, Yibo Ren
arxiv.org/abs/2603.27964 mastoxiv.page/@arXiv_mathDG_bo
- The Hodograph Transform Between Thermodynamics and Relativity
Leonid Polterovich
arxiv.org/abs/2603.28039 mastoxiv.page/@arXiv_grqc_bot/
toXiv_bot_toot

@heiseonline@social.heise.de
2026-04-29 23:30:00

Alphabets Geschäft brummt, Investitionen auch
Der KI-Bot in Googles Suchmaschine führt zu so vielen Eingaben wie noch nie. Das bringt Umsatz. Noch stärker wächst das Cloud-Geschäft.
h…

Alphabets Geschäft brummt, Investitionen auch
Der KI-Bot in Googles Suchmaschine führt zu so vielen Eingaben wie noch nie. Das bringt Umsatz. Noch stärker wächst das Cloud-Geschäft.

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 11:13:03

Replaced article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[4/5]:
- Retrieving Climate Change Disinformation by Narrative
Upravitelev, Solopova, Jakob, Sahitaj, M\"oller, Schmitt
arxiv.org/abs/2603.22015 mastoxiv.page/@arXiv_csCL_bot/
- PaperVoyager : Building Interactive Web with Visual Language Models
Dasen Dai, Biao Wu, Meng Fang, Wenhao Wang
arxiv.org/abs/2603.22999 mastoxiv.page/@arXiv_csCL_bot/
- Continual Robot Skill and Task Learning via Dialogue
Weiwei Gu, Suresh Kondepudi, Anmol Gupta, Lixiao Huang, Nakul Gopalan
arxiv.org/abs/2409.03166 mastoxiv.page/@arXiv_csRO_bot/
- Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs
Zara Siddique, Irtaza Khalid, Liam D. Turner, Luis Espinosa-Anke
arxiv.org/abs/2503.05371 mastoxiv.page/@arXiv_csLG_bot/
- SkillFlow: Scalable and Efficient Agent Skill Retrieval System
Fangzhou Li, Pagkratios Tagkopoulos, Ilias Tagkopoulos
arxiv.org/abs/2504.06188 mastoxiv.page/@arXiv_csAI_bot/
- Large Language Models for Computer-Aided Design: A Survey
Licheng Zhang, Bach Le, Naveed Akhtar, Siew-Kei Lam, Tuan Ngo
arxiv.org/abs/2505.08137 mastoxiv.page/@arXiv_csLG_bot/
- Structured Agent Distillation for Large Language Model
Liu, Kong, Dong, Yang, Li, Tang, Yuan, Niu, Zhang, Zhao, Lin, Huang, Wang
arxiv.org/abs/2505.13820 mastoxiv.page/@arXiv_csLG_bot/
- VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction
Fan, Zhang, Li, Zhang, Chen, Hu, Wang, Qu, Zhou, Wang, Yan, Xu, Theiss, Chen, Li, Tu, Wang, Ranjan
arxiv.org/abs/2505.20279 mastoxiv.page/@arXiv_csCV_bot/
- Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification
Bhattacharjee, Tian, Rubin, Lo, Merchant, Hanson, Gounley, Tandon
arxiv.org/abs/2506.04450 mastoxiv.page/@arXiv_csCR_bot/
- L-MARS: Legal Multi-Agent Workflow with Orchestrated Reasoning and Agentic Search
Ziqi Wang, Boqin Yuan
arxiv.org/abs/2509.00761 mastoxiv.page/@arXiv_csAI_bot/
- Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
Han, Huang, Liao, Jiang, Lu, Zhao, Wang, Zhou, Jiang, Liang, Zhou, Sun, Yu, Xiao
arxiv.org/abs/2509.23392 mastoxiv.page/@arXiv_csAI_bot/
- Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models
Leander Girrbach, Stephan Alaniz, Genevieve Smith, Trevor Darrell, Zeynep Akata
arxiv.org/abs/2510.03721 mastoxiv.page/@arXiv_csCV_bot/
- Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
Zhang, Hu, Upasani, Ma, Hong, Kamanuru, Rainton, Wu, Ji, Li, Thakker, Zou, Olukotun
arxiv.org/abs/2510.04618 mastoxiv.page/@arXiv_csLG_bot/
- Mitigating Premature Exploitation in Particle-based Monte Carlo for Inference-Time Scaling
Giannone, Xu, Nayak, Awhad, Sudalairaj, Xu, Srivastava
arxiv.org/abs/2510.05825 mastoxiv.page/@arXiv_csLG_bot/
- Complete asymptotic type-token relationship for growing complex systems with inverse power-law co...
Pablo Rosillo-Rodes, Laurent H\'ebert-Dufresne, Peter Sheridan Dodds
arxiv.org/abs/2511.02069 mastoxiv.page/@arXiv_physicsso
- ViPRA: Video Prediction for Robot Actions
Sandeep Routray, Hengkai Pan, Unnat Jain, Shikhar Bahl, Deepak Pathak
arxiv.org/abs/2511.07732 mastoxiv.page/@arXiv_csRO_bot/
- AISAC: An Integrated multi-agent System for Transparent, Retrieval-Grounded Scientific Assistance
Chandrachur Bhattacharya, Sibendu Som
arxiv.org/abs/2511.14043
- VideoARM: Agentic Reasoning over Hierarchical Memory for Long-Form Video Understanding
Yufei Yin, Qianke Meng, Minghao Chen, Jiajun Ding, Zhenwei Shao, Zhou Yu
arxiv.org/abs/2512.12360 mastoxiv.page/@arXiv_csCV_bot/
- RadImageNet-VQA: A Large-Scale CT and MRI Dataset for Radiologic Visual Question Answering
L\'eo Butsanets, Charles Corbi\`ere, Julien Khlaut, Pierre Manceron, Corentin Dancette
arxiv.org/abs/2512.17396 mastoxiv.page/@arXiv_csCV_bot/
- Measuring all the noises of LLM Evals
Sida Wang
arxiv.org/abs/2512.21326 mastoxiv.page/@arXiv_csLG_bot/
toXiv_bot_toot

@arXiv_qbioNC_bot@mastoxiv.page
2026-04-30 08:51:16

Replaced article(s) found for q-bio.NC. arxiv.org/list/q-bio.NC/new
[1/1]:
- Untrained CNNs Match Backpropagation at V1: A Systematic RSA Comparison of Four Learning Rules Ag...
Nils Leutenegger
arxiv.org/abs/2604.16875 mastoxiv.page/@arXiv_csLG_bot/
toXiv_bot_toot

@arXiv_nlinAO_bot@mastoxiv.page
2026-03-31 11:18:16

Replaced article(s) found for nlin.AO. arxiv.org/list/nlin.AO/new
[1/1]:
- Elementary derivation of the dissipation--coherence bound for stochastic oscillators
Artemy Kolchinsky
arxiv.org/abs/2510.14101 mastoxiv.page/@arXiv_condmatst
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:40:59

Crosslisted article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[2/2]:
- The Geometry of Harmful Intent: Training-Free Anomaly Detection via Angular Deviation in LLM Resi...
Isaac Llorente-Saguer
arxiv.org/abs/2603.27412 mastoxiv.page/@arXiv_csLG_bot/
- LongCat-Next: Lexicalizing Modalities as Discrete Tokens
Meituan LongCat Team, et al.
arxiv.org/abs/2603.27538 mastoxiv.page/@arXiv_csCV_bot/
- Emergent Social Intelligence Risks in Generative Multi-Agent Systems
Huang, Jiang, Wang, Zhuang, Luo, Ma, Xu, Chen, Moniz, Lin, Chen, Chawla, Dziri, Sun, Zhang
arxiv.org/abs/2603.27771 mastoxiv.page/@arXiv_csMA_bot/
- KVSculpt: KV Cache Compression as Distillation
Bo Jiang, Sian Jin
arxiv.org/abs/2603.27819 mastoxiv.page/@arXiv_csLG_bot/
- Q-Bridge: Code Translation for Quantum Machine Learning via LLMs
Runjia Zeng, Priyabrata Senapati, Ruixiang Tang, Dongfang Liu, Qiang Guan
arxiv.org/abs/2603.27836 mastoxiv.page/@arXiv_quantph_b
- EffiSkill: Agent Skill Based Automated Code Efficiency Optimization
Zimu Wang, Yuling Shi, Mengfan Li, Zijun Liu, Jie M. Zhang, Chengcheng Wan, Xiaodong Gu
arxiv.org/abs/2603.27850 mastoxiv.page/@arXiv_csSE_bot/
- Efficient Inference of Large Vision Language Models
Surendra Pathak
arxiv.org/abs/2603.27960 mastoxiv.page/@arXiv_csLG_bot/
- CDH-Bench: A Commonsense-Driven Hallucination Benchmark for Evaluating Visual Fidelity in Vision-...
Kesheng Chen, Yamin Hu, Qi Zhou, Zhenqian Zhu, Wenjian Luo
arxiv.org/abs/2603.27982 mastoxiv.page/@arXiv_csCV_bot/
- MOSS-VoiceGenerator: Create Realistic Voices with Natural Language Descriptions
Huang, Fan, Jiang, Jiang, Tu, Zhu, Zhang, Zhao, Yang, Fei, Li, Yang, Cheng, Qiu
arxiv.org/abs/2603.28086 mastoxiv.page/@arXiv_csSD_bot/
- Does Claude's Constitution Have a Culture?
Parham Pourdavood
arxiv.org/abs/2603.28123 mastoxiv.page/@arXiv_csCY_bot/
- MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome
Fangda Ye, et al.
arxiv.org/abs/2603.28407 mastoxiv.page/@arXiv_csAI_bot/
- IsoQuant: Hardware-Aligned SO(4) Isoclinic Rotations for LLM KV Cache Compression
Zhongping Ji
arxiv.org/abs/2603.28430 mastoxiv.page/@arXiv_csLG_bot/
- Entropic Claim Resolution: Uncertainty-Driven Evidence Selection for RAG
Davide Di Gioia
arxiv.org/abs/2603.28444 mastoxiv.page/@arXiv_csAI_bot/
- Moving Beyond Review: Applying Language Models to Planning and Translation in Reflection
Seyed Parsa Neshaei, Richard Lee Davis, Tanja K\"aser
arxiv.org/abs/2603.28596 mastoxiv.page/@arXiv_csHC_bot/
- ResAdapt: Adaptive Resolution for Efficient Multimodal Reasoning
Huanxuan Liao, Zhongtao Jiang, Yupu Hao, Yuqiao Tan, Shizhu He, Jun Zhao, Kun Xu, Kang Liu
arxiv.org/abs/2603.28610 mastoxiv.page/@arXiv_csCV_bot/
- The Ultimate Tutorial for AI-driven Scale Development in Generative Psychometrics: Releasing AIGE...
Lara Russell-Lasalandra, Hudson Golino, Luis Eduardo Garrido, Alexander P. Christensen
arxiv.org/abs/2603.28643 mastoxiv.page/@arXiv_csAI_bot/
- SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning
Philip Schroeder, Thomas Weng, Karl Schmeckpeper, Eric Rosen, Stephen Hart, Ondrej Biza
arxiv.org/abs/2603.28730 mastoxiv.page/@arXiv_csRO_bot/
- ParaSpeechCLAP: A Dual-Encoder Speech-Text Model for Rich Stylistic Language-Audio Pretraining
Anuj Diwan, Eunsol Choi, David Harwath
arxiv.org/abs/2603.28737 mastoxiv.page/@arXiv_eessAS_bo
toXiv_bot_toot

@arXiv_csCC_bot@mastoxiv.page
2026-03-31 07:31:07

[2026-03-31 Tue (UTC), 4 new articles found for cs.CC Computational Complexity]
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 11:05:27

Replaced article(s) found for cs.CR. arxiv.org/list/cs.CR/new
[2/2]:
- Combinatorial Privacy: Private Multi-Party Bitstream Grand Sum by Hiding in Birkhoff Polytopes
Praneeth Vepakomma

@arXiv_csCG_bot@mastoxiv.page
2026-03-30 07:31:37

[2026-03-30 Mon (UTC), 1 new article found for cs.CG Computational Geometry]
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 11:12:53

Replaced article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[3/5]:
- Can Small Language Models Handle Context-Summarized Multi-Turn Customer-Service QA? A Synthetic D...
Lakshan Cooray, Deshan Sumanathilaka, Pattigadapa Venkatesh Raju
arxiv.org/abs/2602.00665 mastoxiv.page/@arXiv_csCL_bot/
- SEAD: Self-Evolving Agent for Multi-Turn Service Dialogue
Dai, Gao, Zhang, Wang, Luo, Wang, Wang, Wu, Wang
arxiv.org/abs/2602.03548
- OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering
Yifan Zhu, Xinyu Mu, Tao Feng, Zhonghong Ou, Yuning Gong, Haoran Luo
arxiv.org/abs/2602.03707
- GreekMMLU: A Native-Sourced Multitask Benchmark for Evaluating Language Models in Greek
Zhang, Konomi, Xypolopoulos, Divriotis, Skianis, Nikolentzos, Stamou, Shang, Vazirgiannis
arxiv.org/abs/2602.05150
- Using LLMs for Knowledge Component-level Correctness Labeling in Open-ended Coding Problems
Zhangqi Duan, Arnav Kankaria, Dhruv Kartik, Andrew Lan
arxiv.org/abs/2602.17542 mastoxiv.page/@arXiv_csCL_bot/
- MetaState: Persistent Working Memory Enhances Reasoning in Discrete Diffusion Language Models
Kejing Xia, Mingzhe Li, Lixuan Wei, Zhenbang Du, Xiangchi Yuan, Dachuan Shi, Qirui Jin, Wenke Lee
arxiv.org/abs/2603.01331 mastoxiv.page/@arXiv_csCL_bot/
- A Browser-based Open Source Assistant for Multimodal Content Verification
Milner, Foster, Karmakharm, Razuvayevskaya, Roberts, Porcellini, Teyssou, Bontcheva
arxiv.org/abs/2603.02842 mastoxiv.page/@arXiv_csCL_bot/
- Nw\=ach\=a Mun\=a: A Devanagari Speech Corpus and Proximal Transfer Benchmark for Nepal Bhasha ASR
Sharma, Shrestha, Poudel, Tiwari, Shrestha, Ghimire, Bal
arxiv.org/abs/2603.07554 mastoxiv.page/@arXiv_csCL_bot/
- Model Merging in the Era of Large Language Models: Methods, Applications, and Future Directions
Mingyang Song, Mao Zheng
arxiv.org/abs/2603.09938 mastoxiv.page/@arXiv_csCL_bot/
- AgentDrift: Unsafe Recommendation Drift Under Tool Corruption Hidden by Ranking Metrics in LLM Ag...
Zekun Wu, Adriano Koshiyama, Sahan Bulathwela, Maria Perez-Ortiz
arxiv.org/abs/2603.12564 mastoxiv.page/@arXiv_csCL_bot/
- GhanaNLP Parallel Corpora: Comprehensive Multilingual Resources for Low-Resource Ghanaian Languages
Gyamfi, Azunre, Moore, Budu, Asare, Owusu, Asiamah
arxiv.org/abs/2603.13793 mastoxiv.page/@arXiv_csCL_bot/
- sebis at ArchEHR-QA 2026: How Much Can You Do Locally? Evaluating Grounded EHR QA on a Single Not...
Ibrahim Ebrar Yurt, Fabian Karl, Tejaswi Choppa, Florian Matthes
arxiv.org/abs/2603.13962 mastoxiv.page/@arXiv_csCL_bot/
- ExPosST: Explicit Positioning with Adaptive Masking for LLM-Based Simultaneous Machine Translation
Yuzhe Shang, Pengzhi Gao, Yazheng Yang, Jiayao Ma, Wei Liu, Jian Luan, Jinsong Su
arxiv.org/abs/2603.14903 mastoxiv.page/@arXiv_csCL_bot/
- BanglaSocialBench: A Benchmark for Evaluating Sociopragmatic and Cultural Alignment of LLMs in Ba...
Tanvir Ahmed Sijan, S. M Golam Rifat, Pankaj Chowdhury Partha, Md. Tanjeed Islam, Md. Musfique Anwar
arxiv.org/abs/2603.15949 mastoxiv.page/@arXiv_csCL_bot/
- EngGPT2: Sovereign, Efficient and Open Intelligence
G. Ciarfaglia, et al.
arxiv.org/abs/2603.16430 mastoxiv.page/@arXiv_csCL_bot/
- HypeLoRA: Hyper-Network-Generated LoRA Adapters for Calibrated Language Model Fine-Tuning
Bartosz Trojan, Filip G\k{e}bala
arxiv.org/abs/2603.19278 mastoxiv.page/@arXiv_csCL_bot/
- Automatic Analysis of Collaboration Through Human Conversational Data Resources: A Review
Yi Yu, Maria Boritchev, Chlo\'e Clavel
arxiv.org/abs/2603.19292 mastoxiv.page/@arXiv_csCL_bot/
- Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Langu...
Xinyue Liu, Niloofar Mireshghallah, Jane C. Ginsburg, Tuhin Chakrabarty
arxiv.org/abs/2603.20957 mastoxiv.page/@arXiv_csCL_bot/
- KG-Hopper: Empowering Compact Open LLMs with Knowledge Graph Reasoning via Reinforcement Learning
Shuai Wang, Yinan Yu
arxiv.org/abs/2603.21440 mastoxiv.page/@arXiv_csCL_bot/
toXiv_bot_toot

@arXiv_qbioPE_bot@mastoxiv.page
2026-03-30 09:42:28

Replaced article(s) found for q-bio.PE. arxiv.org/list/q-bio.PE/new
[1/1]:
- Hidden Markov modelling of spatio-temporal dynamics of measles in 1750-1850 Finland
Tiia-Maria Pasanen, Jouni Helske, Tarmo Ketola
arxiv.org/abs/2405.16885 mastoxiv.page/@arXiv_statME_bo
- Beyond cognacy
Gerhard J\"ager
arxiv.org/abs/2507.03005 mastoxiv.page/@arXiv_csCL_bot/
toXiv_bot_toot

@arXiv_physicschemph_bot@mastoxiv.page
2026-03-30 07:49:07

[2026-03-30 Mon (UTC), 7 new articles found for physics.chem-ph Chemical Physics]
toXiv_bot_toot

@45names@mastodon.social
2026-05-29 22:20:03

Patient Nero
bot by @…
No laws were broken in the making of this bot: put the zip ties down.
#satire #potus45

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 11:12:38

Replaced article(s) found for math.LO. arxiv.org/list/math.LO/new
[1/1]:
- Stably Embedded Pairs of Ordered Abelian Groups
Martin Hils, Martina Liccardo, Pierre Touchard
arxiv.org/abs/2308.09989 mastoxiv.page/@arXiv_mathLO_bo
- Generalization of terms via universal algebra
Tommaso Flaminio, Sara Ugolini
arxiv.org/abs/2502.18259 mastoxiv.page/@arXiv_mathLO_bo
- Subvarieties of pointed Abelian l-groups
Filip Jankovec
arxiv.org/abs/2509.05044 mastoxiv.page/@arXiv_mathLO_bo
- Tensor and direct extension of definable subcategories
Mike Prest
arxiv.org/abs/2305.08678 mastoxiv.page/@arXiv_mathRT_bo
- Transfer principles and the Kato-Kuzumaki conjecture
Felipe Gambardella, Konstantinos Kartas
arxiv.org/abs/2603.01815 mastoxiv.page/@arXiv_mathNT_bo
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-31 11:10:45

Replaced article(s) found for math.CT. arxiv.org/list/math.CT/new
[1/1]:
- Relativized universal algebra via partial Horn logic
Yuto Kawase
arxiv.org/abs/2403.19661 mastoxiv.page/@arXiv_mathCT_bo
- Nonabelian $H^2$ with coefficients in a group and with coefficients in a crossed module
Mikhail Borovoi
arxiv.org/abs/1608.07366
- $(\infty,n)$-Limits I: Definition and first consistency results
Lyne Moser, Nima Rasekh, Martina Rovelli
arxiv.org/abs/2312.11101 mastoxiv.page/@arXiv_mathAT_bo
- 2-Functoriality of Initial Semantics, and Applications
Benedikt Ahrens, Ambroise Lafont, Thomas Lamiaux
arxiv.org/abs/2503.10863 mastoxiv.page/@arXiv_csPL_bot/
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 11:13:08

Replaced article(s) found for cs.CL. arxiv.org/list/cs.CL/new
[5/5]:
- AppellateGen: A Benchmark for Appellate Legal Judgment Generation
Yang, Wang, Fan, Hu, Wang, Liu, Zeng, Fu, Gong, Zhang, Li, Zheng, Xu
arxiv.org/abs/2601.01331 mastoxiv.page/@arXiv_csCY_bot/
- Vision-Language Agents for Interactive Forest Change Analysis
James Brock, Ce Zhang, Nantheera Anantrasirichai
arxiv.org/abs/2601.04497 mastoxiv.page/@arXiv_csCV_bot/
- FigEx2: Visual-Conditioned Panel Detection and Captioning for Scientific Compound Figures
Jifeng Song, Arun Das, Pan Wang, Hui Ji, Kun Zhao, Yufei Huang
arxiv.org/abs/2601.08026 mastoxiv.page/@arXiv_csCV_bot/
- Sparse-RL: Breaking the Memory Wall in LLM Reinforcement Learning via Stable Sparse Rollouts
Luo, Zhang, Hu, Zhang, Wang, Su, Sun, Liang, Zhang
arxiv.org/abs/2601.10079 mastoxiv.page/@arXiv_csLG_bot/
- Compounding Disadvantage: Auditing Intersectional Bias in LLM-Generated Explanations Across India...
Amogh Gupta (Neil), Niharika Patil (Neil), Sourojit Ghosh (Neil), SnehalKumar (Neil), S Gaikwad
arxiv.org/abs/2601.14506 mastoxiv.page/@arXiv_csCY_bot/
- Measuring Complexity at the Requirements Stage: Spectral Metrics as Development Effort Predictors
Vierlboeck, Pugliese, Nilchian, Grogan, Babu
arxiv.org/abs/2602.07182 mastoxiv.page/@arXiv_csSE_bot/
- CoPE-VideoLM: Leveraging Codec Primitives For Efficient Video Language Modeling
Sarkar, Pautrat, Miksik, Pollefeys, Armeni, Rad, Dusmanu
arxiv.org/abs/2602.13191 mastoxiv.page/@arXiv_csCV_bot/
- MoD-DPO: Towards Mitigating Cross-modal Hallucinations in Omni LLMs using Modality Decoupled Pref...
Ashutosh Chaubey, Jiacheng Pang, Mohammad Soleymani
arxiv.org/abs/2603.03192 mastoxiv.page/@arXiv_csCV_bot/
- Image Generation Models: A Technical History
Rouzbeh Shirvani
arxiv.org/abs/2603.07455 mastoxiv.page/@arXiv_csCV_bot/
- Rethinking Attention Output Projection: Structured Hadamard Transforms for Efficient Transformers
Shubham Aggarwal, Lokendra Kumar
arxiv.org/abs/2603.08343 mastoxiv.page/@arXiv_csLG_bot/
- FGTR: Fine-Grained Multi-Table Retrieval via Hierarchical LLM Reasoning
Chaojie Sun, Bin Cao, Tiantian Li, Chenyu Hou, Ruizhe Li, Jing Fan
arxiv.org/abs/2603.12702 mastoxiv.page/@arXiv_csIR_bot/
- CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad
Yongqiang Chen, Chenxi Liu, Zhenhao Chen, Tongliang Liu, Bo Han, Kun Zhang
arxiv.org/abs/2603.14575 mastoxiv.page/@arXiv_csLG_bot/
- Silicon Bureaucracy and AI Test-Oriented Education: Contamination Sensitivity and Score Confidenc...
Yiliang Song, Hongjun An, Jiangan Chen, Xuanchen Yan, Huan Song, Jiawei Shao, Xuelong Li
arxiv.org/abs/2603.21636 mastoxiv.page/@arXiv_csAI_bot/
- Problems with Chinchilla Approach 2: Systematic Biases in IsoFLOP Parabola Fits
Eric Czech, Zhiwei Xu, Yael Elmatad, Yixin Wang, William Held
arxiv.org/abs/2603.22339 mastoxiv.page/@arXiv_csLG_bot/
- X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMs
Di Cao, Dongjie Fu, Hai Yu, Siqi Zheng, Xu Tan, Tao Jin
arxiv.org/abs/2603.24596 mastoxiv.page/@arXiv_eessAS_bo
toXiv_bot_toot

@arXiv_mathSG_bot@mastoxiv.page
2026-03-31 11:14:16

Replaced article(s) found for math.SG. arxiv.org/list/math.SG/new
[1/1]:
- Aspherical Lagrangian submanifolds, Audin's conjecture and cyclic dilations
Yin Li
arxiv.org/abs/2308.05086 mastoxiv.page/@arXiv_mathSG_bo
- On Ma\~n\'e's critical value for the two-component Hunter-Saxton system and a infnite dimensional...
Levin Maier
arxiv.org/abs/2503.12901 mastoxiv.page/@arXiv_mathAP_bo
- Homogeneous potentials, Lagrange's identity and Poisson geometry
A. V. Tsiganov
arxiv.org/abs/2511.19903 mastoxiv.page/@arXiv_nlinSI_bo
toXiv_bot_toot

@arXiv_csCC_bot@mastoxiv.page
2026-03-30 07:30:27

[2026-03-30 Mon (UTC), 2 new articles found for cs.CC Computational Complexity]
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 11:05:12

Replaced article(s) found for cs.CR. arxiv.org/list/cs.CR/new
[1/2]:
- Evasion Adversarial Attacks Remain Impractical Against ML-based Network Intrusion Detection Syste...
Mohamed elShehaby, Ashraf Matrawy

@arXiv_nlinAO_bot@mastoxiv.page
2026-03-31 10:53:50

Crosslisted article(s) found for nlin.AO. arxiv.org/list/nlin.AO/new
[1/1]:
- Communication-Induced Bifurcation and Collective Dynamics in Power Packet Networks: A Thermodynam...
Takashi Hikihara
arxiv.org/abs/2603.27446 mastoxiv.page/@arXiv_eessSY_bo
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 09:09:27

Crosslisted article(s) found for math.LO. arxiv.org/list/math.LO/new
[1/1]:
- Do not throw out the baby: Clarithmetics as alternatives to weak arithmetics
Giorgi Japaridze
arxiv.org/abs/2603.26040 mastoxiv.page/@arXiv_csLO_bot/
- On merge-models
Buffi\`ere, Lin, Ne\v{s}et{\v{r}}il, de Mendez}, Siebertz
arxiv.org/abs/2603.26570 mastoxiv.page/@arXiv_csDM_bot/
toXiv_bot_toot

@arXiv_econTH_bot@mastoxiv.page
2026-03-31 10:44:13

Crosslisted article(s) found for econ.TH. arxiv.org/list/econ.TH/new
[1/1]:
- Power Couple? AI Growth and Renewable Energy Investment
Luyi Gui, Tinglong Dai
arxiv.org/abs/2603.26678 mastoxiv.page/@arXiv_csCY_bot/
toXiv_bot_toot

@arXiv_csCG_bot@mastoxiv.page
2026-03-31 07:32:42

[2026-03-31 Tue (UTC), 6 new articles found for cs.CG Computational Geometry]
toXiv_bot_toot

@arXiv_qbioNC_bot@mastoxiv.page
2026-04-28 09:22:12

Crosslisted article(s) found for q-bio.NC. arxiv.org/list/q-bio.NC/new
[1/1]:
- Linear equivalence of nonlinear recurrent neural networks
David G. Clark
arxiv.org/abs/2604.23489 mastoxiv.page/@arXiv_condmatdi
- Robust and Clinically Reliable EEG Biomarkers: A Cross Population Framework for Generalizable Par...
Rasmussen, Wang, Rizk, Pallab, Stuwart, Mancini, Singh, Santosh
arxiv.org/abs/2604.23933 mastoxiv.page/@arXiv_csLG_bot/
- Solution of a large nonlinear recurrent neural network at fixed connectivity
Albert J. Wakhloo
arxiv.org/abs/2604.24141 mastoxiv.page/@arXiv_condmatdi
- From Players to Participants: Citizen Science and Video Games to Understand Cognition
Syrine Salouhou, Edgar Dubourg, Maxwell Scott-Slade, Hugo Spiers, Antoine Coutrot
arxiv.org/abs/2604.24321 mastoxiv.page/@arXiv_csHC_bot/
- Persistent and anti-persistent stride-to-stride fluctuations: an ARFIMA decomposition consistent ...
Philippe Terrier
arxiv.org/abs/2604.24365 mastoxiv.page/@arXiv_qbioQM_bo
- Cortex-Inspired Continual Learning: Unsupervised Instantiation and Recovery of Functional Task Ne...
Kevin McKee, Thomas Hazy, Yicong Zheng, Zacharie Bugaud, Thomas Miconi
arxiv.org/abs/2604.24637 mastoxiv.page/@arXiv_csLG_bot/
- Homology-based Morphometry of Brain Atrophy: Methods and Applications
Donato Quiccione, Mariam Pirashvili, Nathan Broomhead, Sean J. Fallon
arxiv.org/abs/2604.24714 mastoxiv.page/@arXiv_mathAT_bo
toXiv_bot_toot

@45names@mastodon.social
2026-04-30 14:20:04

Needy Amin
bot by @…
No laws were broken in the making of this bot: I wanted to make you smile, not pursue charges.
#satire #potus45

@arXiv_mathCT_bot@mastoxiv.page
2026-03-30 09:07:37

Crosslisted article(s) found for math.CT. arxiv.org/list/math.CT/new
[1/1]:
- The Image of Functor Morphing
Ehud Meir
arxiv.org/abs/2603.26368 mastoxiv.page/@arXiv_mathRT_bo
- The motivic tt-geometry of real quadrics
Jean Paul Schemeil
arxiv.org/abs/2603.26492 mastoxiv.page/@arXiv_mathAG_bo
toXiv_bot_toot

@arXiv_mathSG_bot@mastoxiv.page
2026-03-30 09:34:31

Replaced article(s) found for math.SG. arxiv.org/list/math.SG/new
[1/1]:
- Higher Dimensional Birkhoff attractors (with an appendix by Maxime Zavidovique)
Marie-Claude Arnaud, Vincent Humili\`ere, Claude Viterbo
arxiv.org/abs/2404.00804
- Polynomial integrable systems from cluster structures
Yanpeng Li, Yu Li, Jiang-Hua Lu
arxiv.org/abs/2411.18904 mastoxiv.page/@arXiv_mathSG_bo
- Hyperk\"ahler structures on leaves of hyper-Lie Poisson manifolds
Dadi Ni, Kaichuan Qi
arxiv.org/abs/2502.08979 mastoxiv.page/@arXiv_mathDG_bo
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 10:40:02

Crosslisted article(s) found for cs.CR. arxiv.org/list/cs.CR/new
[1/1]:
- The Finality Calculator: Analyzing and Quantifying Filecoin's Finality Guarantees
Guy Goren, Jorge M. Soares

@arXiv_qbioNC_bot@mastoxiv.page
2026-04-30 07:47:56

[2026-04-30 Thu (UTC), no new articles found for q-bio.NC Neurons and Cognition]
toXiv_bot_toot

@arXiv_nlinAO_bot@mastoxiv.page
2026-03-31 08:13:12

[2026-03-31 Tue (UTC), no new articles found for nlin.AO Adaptation and Self-Organizing Systems]
toXiv_bot_toot

@45names@mastodon.social
2026-05-31 06:20:02

Dozin' Donnie
bot by @…
No laws were broken in the making of this bot: don't make me go to gulag.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:43:17

BitSov: A Composable Bitcoin-Native Architecture for Sovereign Internet Infrastructure
Oliver Aleksander Larsen, Rasmus Thorsen Larsen, Mahyar T. Moghaddam
arxiv.org/abs/2603.28727

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 07:41:57

[2026-03-30 Mon (UTC), 3 new articles found for math.LO Logic]
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-31 07:43:42

[2026-03-31 Tue (UTC), 3 new articles found for math.CT Category Theory]
toXiv_bot_toot

@arXiv_mathSG_bot@mastoxiv.page
2026-03-31 07:46:02

[2026-03-31 Tue (UTC), 1 new article found for math.SG Symplectic Geometry]
toXiv_bot_toot

@arXiv_nlinAO_bot@mastoxiv.page
2026-03-30 07:57:07

[2026-03-30 Mon (UTC), no new articles found for nlin.AO Adaptation and Self-Organizing Systems]
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:41:12

Safeguarding LLMs Against Misuse and AI-Driven Malware Using Steganographic Canaries
Md Raz, Venkata Sai Charan Putrevu, Meet Udeshi, Prashanth Krishnamurthy, Farshad Khorrami, Ramesh Karri
arxiv.org/abs/2603.28655

@45names@mastodon.social
2026-05-31 08:20:03

President TACO
bot by @…
No laws were broken in the making of this bot: don't send me to a camp, please.
#satire #potus45

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 07:41:37

[2026-03-31 Tue (UTC), 11 new articles found for math.LO Logic]
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-30 07:40:17

[2026-03-30 Mon (UTC), no new articles found for math.CT Category Theory]
toXiv_bot_toot

@arXiv_mathSG_bot@mastoxiv.page
2026-03-30 07:42:17

[2026-03-30 Mon (UTC), no new articles found for math.SG Symplectic Geometry]
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:40:47

Interpretable Ensemble Learning for Network Traffic Anomaly Detection: A SHAP-based Explainable AI Framework for Embedded Systems Security
Wanru Shao
arxiv.org/abs/2603.28654

@45names@mastodon.social
2026-05-29 20:20:02

Benedict Donald
bot by @…
No laws were broken in the making of this bot: put the zip ties down.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:39:33

Empowering Mobile Networks Security Resilience by using Post-Quantum Cryptography
Ricardo Alves Faval, Rodrigo Moreira, Fl\'avio de Oliveira Silva
arxiv.org/abs/2603.28626

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 09:06:12

The Grothendieck ring of a non-divisible ordered abelian group is trivial
Blaise Boissonneau, Mathias Stout, Floris Vermeulen
arxiv.org/abs/2603.28483 arxiv.org/pdf/2603.28483 arxiv.org/html/2603.28483
arXiv:2603.28483v1 Announce Type: new
Abstract: We consider the model-theoretic Grothendieck ring of definable sets in ordered abelian groups. It is well-known that $\mathrm{K} \mathbb{Q} \cong \mathbb{Z}[T]/(T^2 T)$ and $\mathrm{K} \mathbb{Z} =0$, but surprisingly little is known about other cases. We present a short computation which shows that they all collapse: $\mathrm{K} G = 0$, unless $G$ is divisible.
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-31 08:02:47

A synthetic construction of universal cocartesian fibrations
Christian Sattler, David W\"arn
arxiv.org/abs/2603.28688 arxiv.org/pdf/2603.28688 arxiv.org/html/2603.28688
arXiv:2603.28688v1 Announce Type: new
Abstract: We give a model-independent construction of directed univalent cocartesian fibrations of $(\infty,1)$-categories, and prove a straightening equivalence against such fibrations. The key step is showing that cocartesian fibrations descend along localisations, which we accomplish by analysing mapping spaces of localisations. Along the way we introduce a directed version of the join construction, giving a sequential colimit description of the full image of any functor.
toXiv_bot_toot

@45names@mastodon.social
2026-05-30 15:20:03

Quarter Flouder
bot by @…
No laws were broken in the making of this bot: please don't ruin my life.
#satire #potus45

@arXiv_mathSG_bot@mastoxiv.page
2026-03-31 07:47:22

Legendrian and Lagrangian higher torsion
Daniel Alvarez Gavela, Kiyoshi Igusa, Michael Sullivan
arxiv.org/abs/2603.28007 arxiv.org/pdf/2603.28007 arxiv.org/html/2603.28007
arXiv:2603.28007v1 Announce Type: new
Abstract: Let $M$ be a closed manifold. We introduce a family of Legendrian isotopy invariants for Legendrians in $J^1M$, which we collectively call Legendrian higher torsion. Given a choice of a class $\mathcal{F}$ of fibre bundles over $M$, equipped with suitable unitary local systems, the Legendrian higher torsion of a Legendrian $\Lambda \subset J^1M$ is the subset of $H^*(M;\mathbf{R})$ consisting of higher Reidemeister torsion cohomology classes of fibre bundles $W$ over $M$ in the class $\mathcal{F}$ such that $\Lambda$ admits a generating function on a stabilization of $W$. For the class of tube bundles in the sense of Waldhausen we call the invariant tube torsion. In particular, we show that the tube torsion of a nearby Lagrangian $L \subset T^*M$ is well-defined when the stable Gauss map $L \to U/O$ is trivial and consists of a union of cosets of a normalized version of the Pontryagin character. We also identify a distinguished coset, invariant under Hamiltonian isotopy of $L$, which we call nearby Lagrangian torsion. We do not know whether nearby Lagrangians must have trivial tube torsion, as would follow from the nearby Lagrangian conjecture. However, we show that there exist Legendrians $\Lambda \subset J^1M$ with nontrivial tube torsion whose projection $\Lambda \to M$ is homotopic to a diffeomorphism.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:34:27

"What Did It Actually Do?": Understanding Risk Awareness and Traceability for Computer-Use Agents
Zifan Peng
arxiv.org/abs/2603.28551

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:55:57

Relational semantics for flat Heyting-Lewis Logic
Jim de Groot, Tadeusz Litak
arxiv.org/abs/2603.28402 arxiv.org/pdf/2603.28402 arxiv.org/html/2603.28402
arXiv:2603.28402v1 Announce Type: new
Abstract: We introduce relational semantics for "flat Heyting-Lewis logic" $\mathsf{HLC}^{\flat}$. This logic arises as the extension of intuitionistic logic with a Lewis-style strict implication modality that, contrary to its "sharp" counterpart $\mathsf{HLC}^{\sharp}$, does not turn meets into joins in its first argument. We prove completeness and the finite model property for $\mathsf{HLC}^{\flat}$ and for several extensions with additional axioms.
toXiv_bot_toot

@arXiv_mathCT_bot@mastoxiv.page
2026-03-31 08:00:47

Higher algebra in $t$-structured tensor triangulated $\infty$-categories
Jiacheng Liang
arxiv.org/abs/2603.27786 arxiv.org/pdf/2603.27786 arxiv.org/html/2603.27786
arXiv:2603.27786v1 Announce Type: new
Abstract: We generalize fundamental notions of higher algebra, traditionally developed within the $\infty$-category of spectra, to the broader setting of $t$-structured tensor triangulated $\infty$-categories ($ttt$-$\infty$-categories). Under a natural structural condition, which we call "projective rigidity", we establish higher categorical analogues of Lazard's theorem and prove the existence and universal property of Cohn localizations. Furthermore, we generalize higher almost ring theory to the $ttt$-$\infty$-categorical setting, showing that $\pi_0$-epimorphic idempotent algebras are in natural bijection with idempotent ideals. By exploiting deformation theory, we establish a general \'etale rigidity theorem, proving that the $\infty$-category of \'etale algebras over a fixed connective base is completely determined by its discrete counterpart. Finally, we characterize the moduli of such projectively rigid $ttt$-$\infty$-categories, demonstrating that the presheaf $\infty$-category on the 1-dimensional framed cobordism $\infty$-category serves as the universal projectively rigid $ttt$-$\infty$-category.
toXiv_bot_toot

@45names@mastodon.social
2026-04-30 05:20:05

Jabba the Gut
bot by @…
No laws were broken in the making of this bot: tell the black helicopters to return to base.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:30:12

Democratizing Federated Learning with Blockchain and Multi-Task Peer Prediction
Leon Witt, Kentaroh Toyoda, Wojciech Samek, Dan Li
arxiv.org/abs/2603.28434

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:02

LombardoGraphia: Automatic Classification of Lombard Orthography Variants
Edoardo Signoroni, Pavel Rychl\'y
arxiv.org/abs/2603.28418 arxiv.org/pdf/2603.28418 arxiv.org/html/2603.28418
arXiv:2603.28418v1 Announce Type: new
Abstract: Lombard, an underresourced language variety spoken by approximately 3.8 million people in Northern Italy and Southern Switzerland, lacks a unified orthographic standard. Multiple orthographic systems exist, creating challenges for NLP resource development and model training. This paper presents the first study of automatic Lombard orthography classification and LombardoGraphia, a curated corpus of 11,186 Lombard Wikipedia samples tagged across 9 orthographic variants, and models for automatic orthography classification. We curate the dataset, processing and filtering raw Wikipedia content to ensure text suitable for orthographic analysis. We train 24 traditional and neural classification models with various features and encoding levels. Our best models achieve 96.06% and 85.78% overall and average class accuracy, though performance on minority classes remains challenging due to data imbalance. Our work provides crucial infrastructure for building variety-aware NLP resources for Lombard.
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:32:52

Arbitrary models of the complete first-order theories of FDZ-rings
Mahmood Sohrabi
arxiv.org/abs/2603.27730 arxiv.org/pdf/2603.27730 arxiv.org/html/2603.27730
arXiv:2603.27730v1 Announce Type: new
Abstract: In this paper, we study arbitrary models of the first-order theory of a ring $A$ where the additive group $A$ is a finitely generated abelian group. Following an earlier paper by this author, Alexei G. Myasnikov and Francis Oger, we call these rings the FDZ-rings or FDZ-algebras. The rings considered are not necessarily unitary, commutative, or associative. We provide criteria for such rings to be quasi finitely axiomatizable (QFA) or bi-interpretable with the ring of integers $\mathbb Z$. We shall also describe all rings elementarily equivalent to such a ring $A$ given certain constraints on $A$.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:30:02

Cryptanalysis of a Lightweight RFID Authentication Protocol Based on a Variable Matrix Encryption Algorithm
Hongjun Wu
arxiv.org/abs/2603.28313

@45names@mastodon.social
2026-05-29 19:20:03

Tangerine Tyrant
bot by @…
No laws were broken in the making of this bot: cancel the hit squad.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:24:27

VulnScout-C: A Lightweight Transformer for C Code Vulnerability Detection
Aymen Lassoued, Nacef Mbarek, Bechir Dardouri, Bassem Ouni, Qing Li, Fakhri Karray
arxiv.org/abs/2603.28309

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:09:47

Not All Subjectivity Is the Same! Defining Desiderata for the Evaluation of Subjectivity in NLP
Urja Khurana, Michiel van der Meer, Enrico Liscio, Antske Fokkens, Pradeep K. Murukannaiah
arxiv.org/abs/2603.28351 arxiv.org/pdf/2603.28351 arxiv.org/html/2603.28351
arXiv:2603.28351v1 Announce Type: new
Abstract: Subjective judgments are part of several NLP datasets and recent work is increasingly prioritizing models whose outputs reflect this diversity of perspectives. Such responses allow us to shed light on minority voices, which are frequently marginalized or obscured by dominant perspectives. It remains a question whether our evaluation practices align with these models' objectives. This position paper proposes seven evaluation desiderata for subjectivity-sensitive models, rooted in how subjectivity is represented in NLP data and models. The desiderata are constructed in a top-down approach, keeping in mind the user-centric impact of such models. We scan the experimental setup of 60 papers and show that various aspects of subjectivity are still understudied: the distinction between ambiguous and polyphonic input, whether subjectivity is effectively expressed to the user, and a lack of interplay between different desiderata, amongst other gaps.
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:28:12

A note on existentially t-henselian fields
Sylvy Anscombe
arxiv.org/abs/2603.27612 arxiv.org/pdf/2603.27612 arxiv.org/html/2603.27612
arXiv:2603.27612v1 Announce Type: new
Abstract: A field is existentially t-henselian if it is has the same existential theory in the first-order language of rings as a field that admits a nontrivial henselian valuation. This property turns out to be equivalent to $\mathbb{Z}$-largeness, which is a property identified in previous work with Fehm, and which holds for $F$ if and only if $tF[\![t]\!]$ is not Diophantine in $F(\!(t)\!)$, without extra constants. In this short note, we further investigate this property in order to count the number of existential theories of henselian valuations on a given field, and to find other characterizations of existential t-henselianity.
toXiv_bot_toot

@45names@mastodon.social
2026-05-31 11:20:02

Paprika Pinochet
bot by @…
No laws were broken in the making of this bot: let's not get carceral over comedy.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:20:08

Evaluating Privilege Usage of Agents on Real-World Tools
Quan Zhang, Lianhang Fu, Lvsi Lian, Gwihwan Go, Yujue Wang, Chijin Zhou, Yu Jiang, Geguang Pu
arxiv.org/abs/2603.28166

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:32

TIEG-Youpu Solution for NeurIPS 2022 WikiKG90Mv2-LSC
Feng Nie, Zhixiu Ye, Sifa Xie, Shuang Wu, Xin Yuan, Liang Yao, Jiazhen Peng, Xu Cheng
arxiv.org/abs/2603.28512 arxiv.org/pdf/2603.28512 arxiv.org/html/2603.28512
arXiv:2603.28512v1 Announce Type: new
Abstract: WikiKG90Mv2 in NeurIPS 2022 is a large encyclopedic knowledge graph. Embedding knowledge graphs into continuous vector spaces is important for many practical applications, such as knowledge acquisition, question answering, and recommendation systems. Compared to existing knowledge graphs, WikiKG90Mv2 is a large scale knowledge graph, which is composed of more than 90 millions of entities. Both efficiency and accuracy should be considered when building graph embedding models for knowledge graph at scale. To this end, we follow the retrieve then re-rank pipeline, and make novel modifications in both retrieval and re-ranking stage. Specifically, we propose a priority infilling retrieval model to obtain candidates that are structurally and semantically similar. Then we propose an ensemble based re-ranking model with neighbor enhanced representations to produce final link prediction results among retrieved candidates. Experimental results show that our proposed method outperforms existing baseline methods and improves MRR of validation set from 0.2342 to 0.2839.
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 07:57:57

Gradualist descriptionalist set theory
David Simmons
arxiv.org/abs/2603.27077 arxiv.org/pdf/2603.27077 arxiv.org/html/2603.27077
arXiv:2603.27077v1 Announce Type: new
Abstract: We introduce a formal language GDST (gradualist descriptionalist set theory) with a family of interpretations indexed by ordinals, as well as a sublanguage NMID (the language of not necessarily monotonic inductive definitions), and show that the assertion that all propositions in NMID have well-defined truth values is equivalent to the existence for each $k \in \mathbb N$ of a sequence of ordinals $\eta_0 < . . . < \eta_k$ such that for each $i < k$, $\eta_i$ is $\eta_{i 1}$-reflecting, a notion we introduce which implies being $\Pi_n$-reflecting for all $n \in \mathbb N$ (and in particular being admissible and recursively Mahlo).
toXiv_bot_toot

@45names@mastodon.social
2026-04-30 10:20:05

Humpty Trumpty
bot by @…
No laws were broken in the making of this bot: no waterboarding please and thank you.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:15:57

Silent Guardians: Independent and Secure Decision Tree Evaluation Without Chatter
Jinyuan Li, Liang Feng Zhang
arxiv.org/abs/2603.28143 arx…

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:12:22

Training data generation for context-dependent rubric-based short answer grading
Pavel \v{S}indel\'a\v{r}, D\'avid Slivka, Christopher Bouma, Filip Pr\'a\v{s}il, Ond\v{r}ej Bojar
arxiv.org/abs/2603.28537 arxiv.org/pdf/2603.28537 arxiv.org/html/2603.28537
arXiv:2603.28537v1 Announce Type: new
Abstract: Every 4 years, the PISA test is administered by the OECD to test the knowledge of teenage students worldwide and allow for comparisons of educational systems. However, having to avoid language differences and annotator bias makes the grading of student answers challenging. For these reasons, it would be interesting to compare methods of automatic student answer grading. To train some of these methods, which require machine learning, or to compute parameters or select hyperparameters for those that do not, a large amount of domain-specific data is needed. In this work, we explore a small number of methods for creating a large-scale training dataset using only a relatively small confidential dataset as a reference, leveraging a set of very simple derived text formats to preserve confidentiality. Using these methods, we successfully created three surrogate datasets that are, at the very least, superficially more similar to the reference dataset than purely the result of prompt-based generation. Early experiments suggest one of these approaches might also lead to improved model training.
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 08:11:02

Ultrahomogeneity and $\omega$-categoricity of monounary algebras
Thomas Quinn-Gregson
arxiv.org/abs/2603.26616 arxiv.org/pdf/2603.26616 arxiv.org/html/2603.26616
arXiv:2603.26616v1 Announce Type: new
Abstract: Ultrahomogeneity and $\omega$-categoricity are two central concepts arising from model theory, with strong connections with oligomorphic permutation groups and quantifier elimination. In particular, both are conditions on the automorphism group of a structure.
The aim of this paper is to describe both the $\omega$-categorical monounary algebras and the ultrahomogeneous monounary algebras of arbitrary cardinalities. We show that a monounary algebra is $\omega$-categorical [ultrahomogeneous] if and only if every element has finite height and Aut$(\mathcal{A})$ has only finitely many 1-orbits [$\mathcal{A}$ is 1-ultrahomogeneous]. Our classification of ultrahomogeneous monounary algebras is then viewed in the context of previously studied variants of ultrahomogeneity, including (partial)-homogeneity and transitivity.
toXiv_bot_toot

@45names@mastodon.social
2026-04-30 15:20:07

Trumpty Dumpty
bot by @…
No laws were broken in the making of this bot: please don't extraordinarily rendition me to a black site.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:14:13

Seeing the Unseen: Rethinking Illicit Promotion Detection with In-Context Learning
Sangyi Wu, Junpu Guo, Xianghang Mi
arxiv.org/abs/2603.28043

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:47

EarlySciRev: A Dataset of Early-Stage Scientific Revisions Extracted from LaTeX Writing Traces
L\'eane Jourdan, Julien Aubert-B\'educhaud, Yannis Chupin, Marah Baccari, Florian Boudin
arxiv.org/abs/2603.28515 arxiv.org/pdf/2603.28515 arxiv.org/html/2603.28515
arXiv:2603.28515v1 Announce Type: new
Abstract: Scientific writing is an iterative process that generates rich revision traces, yet publicly available resources typically expose only final or near-final versions of papers. This limits empirical study of revision behaviour and evaluation of large language models (LLMs) for scientific writing. We introduce EarlySciRev, a dataset of early-stage scientific text revisions automatically extracted from arXiv LaTeX source files. Our key observation is that commented-out text in LaTeX often preserves discarded or alternative formulations written by the authors themselves. By aligning commented segments with nearby final text, we extract paragraph-level candidate revision pairs and apply LLM-based filtering to retain genuine revisions. Starting from 1.28M candidate pairs, our pipeline yields 578k validated revision pairs, grounded in authentic early drafting traces. We additionally provide a human-annotated benchmark for revision detection. EarlySciRev complements existing resources focused on late-stage revisions or synthetic rewrites and supports research on scientific writing dynamics, revision modelling, and LLM-assisted editing.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:06:17

Kill-Chain Canaries: Stage-Level Tracking of Prompt Injection Across Attack Surfaces and Model Safety Tiers
Haochuan Kevin Wang
arxiv.org/abs/2603.28013

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:01:07

Hindman and Owings-like theorems without the Axiom of Choice
Jos\'e A. Guzm\'an-Vega, David J. Fern\'andez Bret\'on, Eliseo Sarmiento Rosales
arxiv.org/abs/2603.27163 arxiv.org/pdf/2603.27163 arxiv.org/html/2603.27163
arXiv:2603.27163v1 Announce Type: new
Abstract: We investigate Hindman- and Owings-type Ramsey-theoretic statements in Zermelo-Fraenkel set theory without the Axiom of Choice, with some occasional extra assumptions (such as the Axiom of Dependent Choice and/or the Axiom of Determinacy). We study several variations of Hindman's theorem on $\mathbb Q$-vector spaces; notably, we show that the uncountable analog of Hindman's theorem fails for the additive group of $\mathbb R$ (under ZF), and for $\mathbb Q$-vector spaces of uncountable dimension (under DC if such dimension is not well-orderable), among other results. In contrast, for Owings-type configurations, we obtain several positive results, especially when assuming AD. These results highlight the interaction between determinacy, algebraic structure, and dimension in the study of infinite Ramsey theory without the Axiom of Choice.
toXiv_bot_toot

@45names@mastodon.social
2026-05-29 17:45:30

Colonnade Crapper
bot by @…
No laws were broken in the making of this bot: let's not have humor lead to homicide.
#satire #potus45

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:12:47

EpiScreen: Early Epilepsy Detection from Electronic Health Records with Large Language Models
Shuang Zhou, Kai Yu, Zaifu Zhan, Huixue Zhou, Min Zeng, Feng Xie, Zhiyi Sha, Rui Zhang
arxiv.org/abs/2603.28698 arxiv.org/pdf/2603.28698 arxiv.org/html/2603.28698
arXiv:2603.28698v1 Announce Type: new
Abstract: Epilepsy and psychogenic non-epileptic seizures often present with similar seizure-like manifestations but require fundamentally different management strategies. Misdiagnosis is common and can lead to prolonged diagnostic delays, unnecessary treatments, and substantial patient morbidity. Although prolonged video-electroencephalography is the diagnostic gold standard, its high cost and limited accessibility hinder timely diagnosis. Here, we developed a low-cost, effective approach, EpiScreen, for early epilepsy detection by utilizing routinely collected clinical notes from electronic health records. Through fine-tuning large language models on labeled notes, EpiScreen achieved an AUC of up to 0.875 on the MIMIC-IV dataset and 0.980 on a private cohort of the University of Minnesota. In a clinician-AI collaboration setting, EpiScreen-assisted neurologists outperformed unaided experts by up to 10.9%. Overall, this study demonstrates that EpiScreen supports early epilepsy detection, facilitating timely and cost-effective screening that may reduce diagnostic delays and avoid unnecessary interventions, particularly in resource-limited regions.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 09:05:12

FedFG: Privacy-Preserving and Robust Federated Learning via Flow-Matching Generation
Ruiyang Wang, Rong Pan, Zhengan Yao
arxiv.org/abs/2603.27986

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:34:52

Determining the normal subgroups of the automorphism groups of some ultrahomogeneous structures via stabilisers
Thomas Bernert, Rob Sullivan, Jeroen Winkel, Shujie Yang
arxiv.org/abs/2603.27890 arxiv.org/pdf/2603.27890 arxiv.org/html/2603.27890
arXiv:2603.27890v1 Announce Type: new
Abstract: We show the simplicity of the automorphism groups of the generic $n$-hypertournament and the semigeneric tournament, and determine the normal subgroups of the automorphism groups of several other ultrahomogeneous oriented graphs. We also give a new proof of the simplicity of the automorphism group of the dense $\frac{2\pi}{n}$-local order $\mathbb{S}(n)$ for $n \geq 2$ (a result due to Droste, Giraudet and Macpherson). Previous techniques of Li, Macpherson, Tent and Ziegler involving stationary weak independence relations (SWIRs) cannot be applied directly to these structures; our approach involves applying these techniques to a certain expansion of each structure, where the expansion has a SWIR and its automorphism group is isomorphic to a stabiliser subgroup of the automorphism group of the original structure.
toXiv_bot_toot

@45names@mastodon.social
2026-05-31 01:20:02

Lil' Donny Moscow
bot by @…
No laws were broken in the making of this bot: let's not have humor lead to homicide.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 08:54:27

Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey
Bhavuk Jain, Sercan \"O. Ar{\i}k, Hardeo K. Thakur
arxiv.org/abs/2603.27918

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:22

Structural-Ambiguity-Aware Translation from Natural Language to Signal Temporal Logic
Kosei Fushimi, Kazunobu Serizawa, Junya Ikemoto, Kazumune Hashimoto
arxiv.org/abs/2603.28426 arxiv.org/pdf/2603.28426 arxiv.org/html/2603.28426
arXiv:2603.28426v1 Announce Type: new
Abstract: Signal Temporal Logic (STL) is widely used to specify timed and safety-critical tasks for cyber-physical systems, but writing STL formulas directly is difficult for non-expert users. Natural language (NL) provides a convenient interface, yet its inherent structural ambiguity makes one-to-one translation into STL unreliable. In this paper, we propose an \textit{ambiguity-preserving} method for translating NL task descriptions into STL candidate formulas. The key idea is to retain multiple plausible syntactic analyses instead of forcing a single interpretation at the parsing stage. To this end, we develop a three-stage pipeline based on Combinatory Categorial Grammar (CCG): ambiguity-preserving $n$-best parsing, STL-oriented template-based semantic composition, and canonicalization with score aggregation. The proposed method outputs a deduplicated set of STL candidates with plausibility scores, thereby explicitly representing multiple possible formal interpretations of an ambiguous instruction. In contrast to existing one-best NL-to-logic translation methods, the proposed approach is designed to preserve attachment and scope ambiguity. Case studies on representative task descriptions demonstrate that the method generates multiple STL candidates for genuinely ambiguous inputs while collapsing unambiguous or canonically equivalent derivations to a single STL formula.
toXiv_bot_toot

@45names@mastodon.social
2026-05-31 10:20:03

Gameshow Göering
bot by @…
No laws were broken in the making of this bot: let's leave the extrajudicial killing out of it.
#satire #potus45…

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 07:59:52

Speedability of computably approximable reals and their approximations
George Barmpalias, Nan Fang, Wolfgang Merkle, Ivan Titov
arxiv.org/abs/2603.26484 arxiv.org/pdf/2603.26484 arxiv.org/html/2603.26484
arXiv:2603.26484v1 Announce Type: new
Abstract: An approximation of a real is a sequence of rational numbers that converges to the real. An approximation is left-c.e. if it is computable and nondecreasing and is d.c.e. if it is computable and has bounded variation. A real is computably approximable if it has some computable approximation, and left-c.e. and d.c.e. reals are defined accordingly.
An approximation $\{a_s\}_{s \in \omega}$ is speedable if there exists a nondecreasing computable function $f$ such that the approximation $\{a_{f(s)}\}_{s \in \omega}$ converges in a certain formal sense faster than $\{a_s\}_{s \in \omega}$. This leads to various notions of speedability for reals, e.g., one may require for a computably approximable real that either all or some of its approximations of a specific type are speedable.
Merkle and Titov established the equivalence of several speedability notions for left-c.e. reals that are defined in terms of left-c.e. approximations. We extend these results to d.c.e. reals and d.c.e. approximations, and we prove that in this setting, being speedable is equivalent to not being Martin-L\"{o}f random. Finally, we demonstrate that every computably approximable real has a computable approximation that is speedable.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 08:54:17

Decentralized Proof-of-Location for Content Provenance: Towards Capture-Time Authenticity
Eduardo Brito, Fernando Castillo, Amnir Hadachi, Ulrich Norbisrath, Jonathan Heiss
arxiv.org/abs/2603.27883

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:27

Courtroom-Style Multi-Agent Debate with Progressive RAG and Role-Switching for Controversial Claim Verification
Masnun Nuha Chowdhury, Nusrat Jahan Beg, Umme Hunny Khan, Syed Rifat Raiyan, Md Kamrul Hasan, Hasan Mahmud
arxiv.org/abs/2603.28488 arxiv.org/pdf/2603.28488 arxiv.org/html/2603.28488
arXiv:2603.28488v1 Announce Type: new
Abstract: Large language models (LLMs) remain unreliable for high-stakes claim verification due to hallucinations and shallow reasoning. While retrieval-augmented generation (RAG) and multi-agent debate (MAD) address this, they are limited by one-pass retrieval and unstructured debate dynamics. We propose a courtroom-style multi-agent framework, PROClaim, that reformulates verification as a structured, adversarial deliberation. Our approach integrates specialized roles (e.g., Plaintiff, Defense, Judge) with Progressive RAG (P-RAG) to dynamically expand and refine the evidence pool during the debate. Furthermore, we employ evidence negotiation, self-reflection, and heterogeneous multi-judge aggregation to enforce calibration, robustness, and diversity. In zero-shot evaluations on the Check-COVID benchmark, PROClaim achieves 81.7% accuracy, outperforming standard multi-agent debate by 10.0 percentage points, with P-RAG driving the primary performance gains ( 7.5 pp). We ultimately demonstrate that structural deliberation and model heterogeneity effectively mitigate systematic biases, providing a robust foundation for reliable claim verification. Our code and data are publicly available at github.com/mnc13/PROClaim.
toXiv_bot_toot

@arXiv_mathLO_bot@mastoxiv.page
2026-03-31 08:06:12

The Cardinalities of Intervals of Equational Theories and Logics
Juan P. Aguilera, Nick Bezhanishvili, Tenyo Takahashi
arxiv.org/abs/2603.27203 arxiv.org/pdf/2603.27203 arxiv.org/html/2603.27203
arXiv:2603.27203v1 Announce Type: new
Abstract: We study the cardinality of classes of equational theories (varieties) and logics by applying descriptive set theory. We affirmatively solve open problems raised by Jackson and Lee [Trans. Am. Math. Soc. 370 (2018), pp. 4785-4812] regarding the cardinalities of subvariety lattices, and by Bezhanishvili et al. [J. Math. Log. (2025), in press] regarding the degrees of the finite model property (fmp). By coding equations and formulas by natural numbers, and theories and logics by real numbers, we examine their position in the Borel hierarchy. We prove that every interval of equational theories in a countable language corresponds to a $\boldsymbol{\Pi}^0_1$ set, and every fmp span of a normal modal logic to a $\boldsymbol{\Pi}^0_2$ set. It follows that they have cardinality either $\leq \aleph_0$ or $2^{\aleph_0}$, provably in ZFC. In the same manner, we observe that the set of pretabular extensions of a tense logic is a $\boldsymbol{\Pi}^0_2$ set, so its cardinality is either $\leq \aleph_0$ or $2^{\aleph_0}$. We also point out a negative solution to another open problem raised by Jackson and Lee [Trans. Am. Math. Soc. 370 (2018), pp. 4785-4812] regarding the existence of independent systems, which relies on Je\v{z}ek et al. [Bull. Aust. Math. Soc. 42 (1990), pp. 57-70].
toXiv_bot_toot

@45names@mastodon.social
2026-04-30 20:24:19

Mustard Melanoma
bot by @…
No laws were broken in the making of this bot: don't have me abducted.
#satire #potus45

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 08:54:07

Ordering Power is Sanctioning Power: Sanction Evasion-MEV and the Limits of On-Chain Enforcement
Di Wu, Yuman Bai, Shoupeng Ren, Xinyu Zhang, Yiyue Cao, Xuechao Wang, Wu Wen, Jian Liu
arxiv.org/abs/2603.27739

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:12:07

Compressing Transformer Language Models via Matrix Product Operator Decomposition: A Case Study on PicoGPT
Younes Javanmard, Tanmoy Pandit, Masoud Mardani
arxiv.org/abs/2603.28534 arxiv.org/pdf/2603.28534 arxiv.org/html/2603.28534
arXiv:2603.28534v1 Announce Type: new
Abstract: Transformer-based language models achieve strong performance across NLP tasks, but their quadratic parameter scaling with hidden dimension makes deployment on resource-constrained hardware expensive. We study Matrix Product Operator (MPO) decomposition as a principled compression method for transformers. MPO factorises weight matrices into chains of low-rank cores, with approximation quality controlled by the bond dimension chi. We replace every nn.Linear layer in PicoGPT, a GPT-2-style character-level language model with about 1M parameters, with an MPOLinear module parameterised as an MPO chain. Cores are initialised either by TT-SVD from pretrained dense weights or from random initialisation, and trained using standard PyTorch autograd without a custom backward pass. We derive balanced factorisation schemes for the five distinct weight shapes in PicoGPT and evaluate bond dimensions chi in {4, 8, 16, 32} on Tiny Shakespeare. MPO compression achieves up to 13x compression per transformer block at chi = 4. At chi = 16, the model uses 191,872 parameters instead of 1,020,224 while retaining 97.7% of baseline token accuracy (51.6% vs 52.8%). Reconstruction error follows the expected trend and is lower for three-site than two-site factorisations at the same bond dimension. The chi = 8 model gives the best accuracy per parameter, exceeding the dense baseline by 2.7x on this metric. These results show that MPO parameterisation is a practical and theoretically grounded alternative to low-rank methods and unstructured pruning for transformer compression.
toXiv_bot_toot

@arXiv_csCR_bot@mastoxiv.page
2026-03-31 08:51:57

A Systematic Taxonomy of Security Vulnerabilities in the OpenClaw AI Agent Framework
Surada Suwansathit, Yuxuan Zhang, Guofei Gu
arxiv.org/abs/2603.27517 …

@arXiv_mathLO_bot@mastoxiv.page
2026-03-30 07:53:07

A One-Step Cascade Symmetric Model: Rank-$1$ Packets, Binary Shielding, and the Even Exact-Cardinality Profile
Frank Gilson
arxiv.org/abs/2603.25950 arxiv.org/pdf/2603.25950 arxiv.org/html/2603.25950
arXiv:2603.25950v1 Announce Type: new
Abstract: We introduce a one-step cascade symmetric system whose local symmetry geometry is organized by finite $\rho$-closed windows and one-step stars rather than by rowwise-independent toggles. The resulting symmetric model isolates a new $ZF DC \neg \mathrm{BPI}$ geometry in which rank-$1$ hereditarily symmetric reals admit a packet normalization theorem over countable $\rho$-closed supports.
The technical center of the paper is the finite star-span lemma and the associated rank-$1$ packet calculus. From this we obtain a normalization theorem and a two-layer coding consequence for rank-$1$ reals (in the metatheory, via a well-orderable base of packets). We then apply the same binary fresh-support shielding pattern to prove $\neg C_2$, hence $\neg AC_{\mathrm{fin}}$, and therefore the failure of every even $C_n$ (where $C_n$ denotes the principle that every family of nonempty $n$-element sets admits a choice function). On the odd side, the present bounded packet calculus remains dyadic: support-fixed local actions factor through finite $2$-groups, bounded support-equivariant quotients of finite local orbits have power-of-two size, and trace-separated bounded rigid ternary families admit canonical selectors within a fixed finite trace window. Accordingly, the odd exact-cardinality profile remains open beyond the current local binary machinery.
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:13:27

Adaptive Block-Scaled Data Types
Jack Cook, Hyemin S. Lee, Kathryn Le, Junxian Guo, Giovanni Traverso, Anantha P. Chandrakasan, Song Han
arxiv.org/abs/2603.28765 arxiv.org/pdf/2603.28765 arxiv.org/html/2603.28765
arXiv:2603.28765v1 Announce Type: new
Abstract: NVFP4 has grown increasingly popular as a 4-bit format for quantizing large language models due to its hardware support and its ability to retain useful information with relatively few bits per parameter. However, the format is not without limitations: recent work has shown that NVFP4 suffers from its error distribution, resulting in large amounts of quantization error on near-maximal values in each group of 16 values. In this work, we leverage this insight to design new Adaptive Block-Scaled Data Types that can adapt to the distribution of their input values. For four-bit quantization, our proposed IF4 (Int/Float 4) data type selects between FP4 and INT4 representations for each group of 16 values, which are then scaled by an E4M3 scale factor as is done with NVFP4. The selected data type is denoted using the scale factor's sign bit, which is currently unused in NVFP4, and we apply the same insight to design formats for other bit-widths, including IF3 and IF6. When used to quantize language models, we find that IF4 outperforms existing 4-bit block-scaled formats, achieving lower loss during quantized training and achieving higher accuracy on many tasks in post-training quantization. We additionally design and evaluate an IF4 Multiply-Accumulate (MAC) unit to demonstrate that IF4 can be implemented efficiently in next-generation hardware accelerators. Our code is available at github.com/mit-han-lab/fourove.
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:09:52

Tailoring AI-Driven Reading Scaffolds to the Distinct Needs of Neurodiverse Learners
Soufiane Jhilal, Eleonora Pasqua, Caterina Marchesi, Riccardo Corradi, Martina Galletti
arxiv.org/abs/2603.28370 arxiv.org/pdf/2603.28370 arxiv.org/html/2603.28370
arXiv:2603.28370v1 Announce Type: new
Abstract: Neurodiverse learners often require reading supports, yet increasing scaffold richness can sometimes overload attention and working memory rather than improve comprehension. Grounded in the Construction-Integration model and a contingent scaffolding perspective, we examine how structural versus semantic scaffolds shape comprehension and reading experience in a supervised inclusive context. Using an adapted reading interface, we compared four modalities: unmodified text, sentence-segmented text, segmented text with pictograms, and segmented text with pictograms plus keyword labels. In a within-subject pilot with 14 primary-school learners with special educational needs and disabilities, we measured reading comprehension using standardized questions and collected brief child- and therapist-reported experience measures alongside open-ended feedback. Results highlight heterogeneous responses as some learners showed patterns consistent with benefits from segmentation and pictograms, while others showed patterns consistent with increased coordination costs when visual scaffolds were introduced. Experience ratings showed limited differences between modalities, with some apparent effects linked to clinical complexity, particularly for perceived ease of understanding. Open-ended feedback of the learners frequently requested simpler wording and additional visual supports. These findings suggest that no single scaffold is universally optimal, reinforcing the need for calibrated, adjustable scaffolding and provide design implications for human-AI co-regulation in supervised inclusive reading contexts.
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:10:07

Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design
Bin Zhu, Qianghuai Jia, Tian Lan, Junyang Ren, Feng Gu, Feihu Jiang, Longyue Wang, Zhao Xu, Weihua Luo
arxiv.org/abs/2603.28376 arxiv.org/pdf/2603.28376 arxiv.org/html/2603.28376
arXiv:2603.28376v1 Announce Type: new
Abstract: Deep research agents autonomously conduct open-ended investigations, integrating complex information retrieval with multi-step reasoning across diverse sources to solve real-world problems. To sustain this capability on long-horizon tasks, reliable verification is critical during both training and inference. A major bottleneck in existing paradigms stems from the lack of explicit verification mechanisms in QA data synthesis, trajectory construction, and test-time scaling. Errors introduced at each stage propagate downstream and degrade the overall agent performance. To address this, we present Marco DeepResearch, a deep research agent optimized with a verification-centric framework design at three levels: \textbf{(1)~QA Data Synthesis:} We introduce verification mechanisms to graph-based and agent-based QA synthesis to control question difficulty while ensuring answers are unique and correct; \textbf{(2)~Trajectory Construction:} We design a verification-driven trajectory synthesis method that injects explicit verification patterns into training trajectories; and \textbf{(3)~Test-time scaling:} We use Marco DeepResearch itself as a verifier at inference time and effectively improve performance on challenging questions. Extensive experimental results demonstrate that our proposed Marco DeepResearch agent significantly outperforms 8B-scale deep research agents on most challenging benchmarks, such as BrowseComp and BrowseComp-ZH. Crucially, under a maximum budget of 600 tool calls, Marco DeepResearch even surpasses or approaches several 30B-scale agents, like Tongyi DeepResearch-30B.
toXiv_bot_toot

@arXiv_csCL_bot@mastoxiv.page
2026-03-31 10:11:57

GraphWalker: Agentic Knowledge Graph Question Answering via Synthetic Trajectory Curriculum
Shuwen Xu, Yao Xu, Jiaxiang Liu, Chenhao Yuan, Wenshuo Peng, Jun Zhao, Kang Liu
arxiv.org/abs/2603.28533 arxiv.org/pdf/2603.28533 arxiv.org/html/2603.28533
arXiv:2603.28533v1 Announce Type: new
Abstract: Agentic knowledge graph question answering (KGQA) requires an agent to iteratively interact with knowledge graphs (KGs), posing challenges in both training data scarcity and reasoning generalization. Specifically, existing approaches often restrict agent exploration: prompting-based methods lack autonomous navigation training, while current training pipelines usually confine reasoning to predefined trajectories. To this end, this paper proposes \textit{GraphWalker}, a novel agentic KGQA framework that addresses these challenges through \textit{Automated Trajectory Synthesis} and \textit{Stage-wise Fine-tuning}. GraphWalker adopts a two-stage SFT training paradigm: First, the agent is trained on structurally diverse trajectories synthesized from constrained random-walk paths, establishing a broad exploration prior over the KG; Second, the agent is further fine-tuned on a small set of expert trajectories to develop reflection and error recovery capabilities. Extensive experiments demonstrate that our stage-wise SFT paradigm unlocks a higher performance ceiling for a lightweight reinforcement learning (RL) stage, enabling GraphWalker to achieve state-of-the-art performance on CWQ and WebQSP. Additional results on GrailQA and our constructed GraphWalkerBench confirm that GraphWalker enhances generalization to out-of-distribution reasoning paths. The code is publicly available at github.com/XuShuwenn/GraphWalk
toXiv_bot_toot