Tootfinder

New bounds for some small multicolor Ramsey numbers
William J. Wesley
https://arxiv.org/abs/2509.03784 https://arxiv.org/pdf/2509.03784

New bounds for some small multicolor Ramsey numbers
The Ramsey number $R(G_1,\dots,G_k)$ is the smallest $n$ such that every $k$-coloring of the edges of $K_n$ contains a monochromatic copy of $G_i$ in color $i$. Ramsey numbers are challenging to compute, and few are known exactly. We use Boolean satisfiability (SAT) solvers to search for structured colorings that give lower bounds, and we show $R(K_4,K_4-e,K_4-e) \ge 35$ and $R(K_3,K_4,C_4,C_4) \ge 49$. Moreover, we tighten some recent upper bounds for multicolor Ramsey numbers for cycles and s…

5 reasons to submit your idea for #TNC26 Call for Proposals - Part 1
Next June, Helsinki 🇫🇮 will welcome our global R&E community under the theme Digital Sisu—inspired by the Finnish concept of “sisu:” inner strength, perseverance, and determination when things get tough.
In our digital era, sisu is about taking action, meeting challenges with courage, creativity, and collaboration to…

Eagles pass game not on 'same page' in 1st loss https://www.espn.com/nfl/story/_/id/46502665/devonta-smith-eagles-pass-game-not-same-page-1st-loss

PG-Agent: An Agent Powered by Page Graph
Weizhi Chen, Ziwei Wang, Leyang Yang, Sheng Zhou, Xiaoxuan Tang, Jiajun Bu, Yong Li, Wei Jiang
https://arxiv.org/abs/2509.03536 https://…

PG-Agent: An Agent Powered by Page Graph
Graphical User Interface (GUI) agents possess significant commercial and social value, and GUI agents powered by advanced multimodal large language models (MLLMs) have demonstrated remarkable potential. Currently, existing GUI agents usually utilize sequential episodes of multi-step operations across pages as the prior GUI knowledge, which fails to capture the complex transition relationship between pages, making it challenging for the agents to deeply perceive the GUI environment and generaliz…

I didn't know I shared a birthday with so many famous people. Will Rogers. John Basilone. Walter Cronkite. Loretta Swit.
I wonder if anyone would notice if I added myself to the Wikipedia page for today?
https://en.wikipedia.org/wiki/November_4

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/4]:
- Test-Time Defense Against Adversarial Attacks via Stochastic Resonance of Latent Ensembles
Dong Lao, Yuxiang Zhang, Haniyeh Ehsani Oskouie, Yangchao Wu, Alex Wong, Stefano Soatto

edit_wikiquote: Wikiquote edits (2010)
A bipartite user-page network extracted from Wikiquotes. A user connects to a page if that user edited that page. Edits (edges) are timestamped. Edge weights represent counts of the number of edits.
This network has 423 nodes and 667 edges.
Tags: Informational, Web graph, Multigraph, Timestamps

Elliptical Wilson loops in ${\cal N}=4$ Super Yang-Mills
Thales Azevedo, Alfonso Ballon-Bayona, Leonardo Pazito
https://arxiv.org/abs/2509.04355 https://ar…

Elliptical Wilson loops in ${\cal N}=4$ Super Yang-Mills
We investigate elliptical Wilson loops in ${\cal N}=4$ Super Yang--Mills theory at weak and strong coupling for small values of the eccentricity. We obtain analytical results for the vacuum expectation value of the Wilson loop in the form of a series in the eccentricity parameter. At weak coupling, we use perturbation theory in ${\cal N}=4$ Super Yang--Mills. At strong coupling, we use the AdS/CFT correspondence, which maps the Wilson loop to the minimal-area worldsheet of an open string in AdS…

Team Xiaomi EV-AD VLA: Caption-Guided Retrieval System for Cross-Modal Drone Navigation - Technical Report for IROS 2025 RoboSense Challenge Track 4
Lingfeng Zhang, Erjia Xiao, Yuchen Zhang, Haoxiang Fu, Ruibin Hu, Yanbiao Ma, Wenbo Ding, Long Chen, Hangjun Ye, Xiaoshuai Hao
https://arxiv.org/abs/2510.02728…

Team Xiaomi EV-AD VLA: Caption-Guided Retrieval System for Cross-Modal Drone Navigation - Technical Report for IROS 2025 RoboSense Challenge Track 4
Cross-modal drone navigation remains a challenging task in robotics, requiring efficient retrieval of relevant images from large-scale databases based on natural language descriptions. The RoboSense 2025 Track 4 challenge addresses this challenge, focusing on robust, natural language-guided cross-view image retrieval across multiple platforms (drones, satellites, and ground cameras). Current baseline methods, while effective for initial retrieval, often struggle to achieve fine-grained semantic…

The linear minimal 4-chart with three crossings
Teruo Nagase, Akiko Shima
https://arxiv.org/abs/2509.04114 https://arxiv.org/pdf/2509.04114

The linear minimal 4-chart with three crossings
Charts are oriented labeled graphs in a disk. Any simple surface braid (2-dimensional braid) can be described by using a chart. Also, a chart represents an oriented closed surface embedded in 4-space. In this paper, we investigate embedded surfaces in 4-space by using charts. Let $Γ$ be a chart, and we denote by $Cross(Γ)$ the set of all the crossings of $Γ$, and we denote by $Γ_m$ the union of all the edges of label $m$. For a 4-chart $Γ$, if the closure of each connected component of t…

I run a gloriously dysfunctional website over at Codeberg, where the HTML cries itself to sleep every night:
You can check it out here: #WebsiteDevelopment…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[4/4]:
- Mind the Gap: The Divergence Between Human and LLM-Generated Tasks
Yi-Long Lu, Jiajun Song, Chunhui Zhang, Wei Wang

Jimmy Wales says Wikipedia's "Gaza genocide" page failed to meet its standards of neutrality; the article is listed as "protected" until 21:47 UTC on November 4 (Xander Elliards/The National)
https://www.thenational.scot/news/25591165

Wikipedia row erupts as Jimmy Wales intervenes on 'Gaza genocide' page
WIKIPEDIA founder Jimmy Wales has intervened over the platform’s article on Gaza genocide ...

Factory Farm Alert: 1.2 Million Birds to Be Imprisoned and Killed #AnimalRights

Factory Farm Alert: 1.2 Million Birds to Be Imprisoned and Killed
Please sign PETA’s petition to help stop these plans and spare millions of lives.

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[4/6]:
- BSMamba: Brightness and Semantic Modeling for Long-Range Interaction in Low-Light Image Enhancement
Tongshun Zhang, Pingping Liu, Mengen Cai, Zijian Zhang, Yubing Lu, Qiuzhan Zhou

The symmetric groups $S_n, n\geq 4$, and finite non-abelian simple groups are not embeddable in any Riordan group
Tian-Xiao He, Nikolai A. Krylov
https://arxiv.org/abs/2508.03056

The symmetric groups $S_n, n\geq 4$, and finite non-abelian simple groups are not embeddable in any Riordan group
We prove that the symmetric group of degree greater than three cannot be embedded into the Riordan group with coefficients in any commutative ring. We also prove the impossibility to embed finite non-abelian simple groups. As a closely related topic, we show why all truncated Riordan groups are solvable, in stark contrast to the unsolvability of the infinite-sized Riordan groups. Finally, we give an explicit embedding of the alternating group $A_4$ into the Lagrange subgroup with coefficients i…

Very interesting looking paper: In tech we trust: A history of technophilia in the Intergovernmental Panel on Climate Change's (IPCC) climate mitigation expertise
https://www.sciencedirect.com/science/article/pii/S2214629625003615
Hopefully someone wil…

LLM-Generated Samples for Android Malware Detection
Nik Rollinson, Nikolaos Polatidis
https://arxiv.org/abs/2510.02391 https://arxiv.org/pdf/2510.02391

LLM-Generated Samples for Android Malware Detection
Android malware continues to evolve through obfuscation and polymorphism, posing challenges for both signature-based defenses and machine learning models trained on limited and imbalanced datasets. Synthetic data has been proposed as a remedy for scarcity, yet the role of large language models (LLMs) in generating effective malware data for detection tasks remains underexplored. In this study, we fine-tune GPT-4.1-mini to produce structured records for three malware families: BankBot, Locker/SL…

J'ai trouvé #pedantix nº1241 en 24 coups !
🟩🟩🟩🟩🟩🟩🟩🟩🟩🟧🟧🟧🟥🟥🟥🟥🟥🟥🟥🟥
https://pedantix.certitudes.org/

pédantix - Découvrez la page Wikipédia !
Découvrez la page Wikipédia en devinant les mots cachés.

SmartLLMs Scheduler: A Framework for Cost-Effective LLMs Utilization
Yueyue Liu, Hongyu Zhang, Yuantian Miao
https://arxiv.org/abs/2508.03258 https://arxiv…

SmartLLMs Scheduler: A Framework for Cost-Effective LLMs Utilization
Large Language Models (LLMs) such as GPT-4 and Llama have shown remarkable capabilities in a variety of software engineering tasks. Despite the advancements, their practical deployment faces challenges, including high financial costs, long response time, and varying performance, especially when handling a large number of queries (jobs). Existing optimization strategies for deploying LLMs for diverse tasks focus on static scheduling, which requires extensive training data for performance predict…

GWTC-4.0: Constraints on the Cosmic Expansion Rate and Modified Gravitational-wave Propagation
The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration
https://arxiv.org/abs/2509.04348

GWTC-4.0: Constraints on the Cosmic Expansion Rate and Modified Gravitational-wave Propagation
We analyze data from 142 of the 218 gravitational-wave (GW) sources in the fourth LIGO-Virgo-KAGRA Collaboration (LVK) Gravitational-Wave Transient Catalog (GWTC-4.0) to estimate the Hubble constant $H_0$ jointly with the population properties of merging compact binaries. We measure the luminosity distance and redshifted masses of GW sources directly; in contrast, we infer GW source redshifts statistically through i) location of features in the compact object mass spectrum and merger rate evolu…

Polarization-Controlled Dual-State Distribution of Bell and N00N Entanglement Over a Metro-Scale Commercial Quantum Network
Kazi Reaz, Md Mehdi Hassan, Jacob E. Humberd, Matthew L. Boone, Angel Fraire Estrada, Rick Mukherjee, H. R. Sadeghpour, Girish S. Agarwal, George Siopsis, Tian Li
https://arxiv.org/abs/2509.03701

Polarization-Controlled Dual-State Distribution of Bell and N00N Entanglement Over a Metro-Scale Commercial Quantum Network
We report the first demonstration of multiphoton, dual state entanglement distribution over a metropolitan scale commercial fiber network, implemented on the EPB Bohr IV quantum network in Chattanooga, TN, using an all fiber optic experimental platform. Employing a spatially degenerate, continuous wave type II SPDC biphoton source and fully fiber coupled linear optics, we generated a 4 photon entangled state. Through polarization projective measurements on two locally retained photons, we then …

5 reasons to submit your idea for #TNC26 Call for Proposals - Part 2
Next June, Helsinki 🇫🇮 will welcome our global R&E community under the theme Digital Sisu—inspired by the Finnish concept of “sisu:” inner strength, perseverance, and determination when things get tough.
In our digital era, sisu is about taking action, meeting challenges with courage, creativity, and collaboration to…

edit_wikiquote: Wikiquote edits (2010)
A bipartite user-page network extracted from Wikiquotes. A user connects to a page if that user edited that page. Edits (edges) are timestamped. Edge weights represent counts of the number of edits.
This network has 550 nodes and 938 edges.
Tags: Informational, Web graph, Multigraph, Timestamps

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[4/4]:
- Pilot Study on Generative AI and Critical Thinking in Higher Education Classrooms
W. F. Lamberti, S. R. Lawrence, D. White, S. Kim, S. Abdullah

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/6]:
- C2AL: Cohort-Contrastive Auxiliary Learning for Large-scale Recommendation Systems
Cokbas, Liu, Tao, Veliz, Huang, Wen, Li, Jin, Duman, Au, Lebanon, Chordia, Zhang

RadProPoser: A Framework for Human Pose Estimation with Uncertainty Quantification from Raw Radar Data
Jonas Leo Mueller, Lukas Engel, Eva Dorschky, Daniel Krauss, Ingrid Ullmann, Martin Vossiek, Bjoern M. Eskofier
https://arxiv.org/abs/2508.03578

RadProPoser: A Framework for Human Pose Estimation with Uncertainty Quantification from Raw Radar Data
Radar-based human pose estimation (HPE) provides a privacy-preserving, illumination-invariant sensing modality but is challenged by noisy, multipath-affected measurements. We introduce RadProPoser, a probabilistic encoder-decoder architecture that processes complex-valued radar tensors from a compact 3-transmitter, 4-receiver MIMO radar. By incorporating variational inference into keypoint regression, RadProPoser jointly predicts 26 three-dimensional joint locations alongside heteroscedastic al…

FocusAgent: Simple Yet Effective Ways of Trimming the Large Context of Web Agents
Imene Kerboua, Sahar Omidi Shayegan, Megh Thakkar, Xing Han L\`u, L\'eo Boisvert, Massimo Caccia, J\'er\'emy Espinas, Alexandre Aussem, V\'eronique Eglin, Alexandre Lacoste
https://arxiv.org/abs/2510.03204…

FocusAgent: Simple Yet Effective Ways of Trimming the Large Context of Web Agents
Web agents powered by large language models (LLMs) must process lengthy web page observations to complete user goals; these pages often exceed tens of thousands of tokens. This saturates context limits and increases computational cost processing; moreover, processing full pages exposes agents to security risks such as prompt injection. Existing pruning strategies either discard relevant content or retain irrelevant context, leading to suboptimal action prediction. We introduce FocusAgent, a sim…

Page curves and island's delays in asymptotically flat 2d spacetimes with injections
Yuuta Saito
https://arxiv.org/abs/2509.03997 https://arxiv.org/pdf…

Page curves and island's delays in asymptotically flat 2d spacetimes with injections
We explore spacetimes with multiple energy injections in asymptotically flat two-dimensional black hole and analyze the entanglement entropy in such spacetimes. This work is an extension of the setup of the single-injection case, by F.~F.~Gautason, L.~Schneiderbauer, W.~Sybesma and L.~Thorlacius, to include the multiple energy injections. We derive the solution of the model, by J.~G.~Russo, L.~Susskind and L.~Thorlacius, for a general number $n$ of the total injections, and discuss the entropy …

Selmer-Inspired Elliptic Curve Generation
Awnon Bhowmik
https://arxiv.org/abs/2510.02383 https://arxiv.org/pdf/2510.02383

Selmer-Inspired Elliptic Curve Generation
Elliptic curve cryptography (ECC) is foundational to modern secure communication, yet existing standard curves have faced scrutiny for opaque parameter-generation practices. This work introduces a Selmer-inspired framework for constructing elliptic curves that is both transparent and auditable. Drawing from $2$- and $3$-descent methods, we derive binary quartics and ternary cubics whose classical invariants deterministically yield candidate $(c_4,c_6)$ parameters. Local solubility checks, model…

Low-Cost Open-Source Ambidextrous Robotic Hand with 23 Direct-Drive servos for American Sign Language Alphabet
Kelvin Daniel Gonzalez Amador
https://arxiv.org/abs/2509.03690 htt…

Low-Cost Open-Source Ambidextrous Robotic Hand with 23 Direct-Drive servos for American Sign Language Alphabet
Accessible communication through sign language is vital for deaf communities, 1 yet robotic solutions are often costly and limited. This study presents VulcanV3, a low- 2 cost, open-source, 3D-printed ambidextrous robotic hand capable of reproducing the full 3 American Sign Language (ASL) alphabet (52 signs for right- and left-hand configurations). 4 The system employs 23 direct-drive servo actuators for precise finger and wrist movements, 5 controlled by an Arduino Mega with dual PCA9685 modul…

edit_wikiquote: Wikiquote edits (2010)
A bipartite user-page network extracted from Wikiquotes. A user connects to a page if that user edited that page. Edits (edges) are timestamped. Edge weights represent counts of the number of edits.
This network has 2355 nodes and 9143 edges.
Tags: Informational, Web graph, Multigraph, Timestamps

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[3/4]:
- Neural Jump ODEs as Generative Models
Robert A. Crowell, Florian Krach, Josef Teichmann

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[4/5]:
- A Survey of Pun Generation: Datasets, Evaluations and Methodologies
Yuchen Su, Yonghua Zhu, Ruofan Wang, Zijian Huang, Diana Benavides-Prado, Michael Witbrock

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[3/4]:
- LaMPE: Length-aware Multi-grained Positional Encoding for Adaptive Long-context Scaling Without T...
Sikui Zhang, Guangze Gao, Ziyun Gan, Chunfeng Yuan, Zefeng Lin, Houwen Peng, Bing Li, Weiming Hu

edit_wikinews: Wikipedia news edits (2010)
Two bipartite user-page networks extracted from Wikipedia, about news events. A user connects to a page if that user edited that page. Edits (edges) are timestamped. Edge weights represent counts of the number of edits.
This network has 61822 nodes and 686017 edges.
Tags: Informational, Web graph, Multigraph, Timestamps

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[2/4]:
- Adaptive Deception Framework with Behavioral Analysis for Enhanced Cybersecurity Defense
Basil Abdullah AL-Zahrani

Crosslisted article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[4/7]:
- Knowledge-Graph Based RAG System Evaluation Framework
Sicheng Dong, Vahid Zolfaghari, Nenad Petrovic, Alois Knoll

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[2/4]:
- MetaGen Blended RAG: Unlocking Zero-Shot Precision for Specialized Domain Question-Answering
Kunal Sawarkar, Shivam R. Solanki, Abhilasha Mangal

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/4]:
- EEFSUVA: A New Mathematical Olympiad Benchmark
Nicole N Khatibi, Daniil A. Radamovich, Michael P. Brenner

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[3/4]:
- Transferable Mask Transformer: Cross-domain Semantic Segmentation with Region-adaptive Transferab...
Jianhua Liu, Zhengyu Li, Yanru Wu, Jingge Wang, Yang Tan, Ruizhe Zhao, Guan Wang, Yang Li

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[1/4]:
- Pre-trained Transformer-Based Approach for Arabic Question Answering : A Comparative Study
Kholoud Alsubhi, Amani Jamal, Areej Alhothali

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/5]:
- Large Language Models for Cryptocurrency Transaction Analysis: A Bitcoin Case Study
Lei, Xiang, Wang, Dowsley, Yuen, Choo, Yu

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[2/4]:
- FFHFlow: Diverse and Uncertainty-Aware Dexterous Grasp Generation via Flow Variational Inference
Qian Feng, Jianxiang Feng, Zhaopeng Chen, Rudolph Triebel, Alois Knoll

Transcribe, Translate, or Transliterate: An Investigation of Intermediate Representations in Spoken Language Models
Tol\'ul\d{o}p\'e \`Og\'unr\`em\'i, Christopher D. Manning, Dan Jurafsky, Karen Livescu
https://arxiv.org/abs/2510.02569

Transcribe, Translate, or Transliterate: An Investigation of Intermediate Representations in Spoken Language Models
Spoken language models (SLMs) that integrate speech with large language models (LMs) rely on modality adapters (MAs) to map the output of speech encoders to a representation that is understandable to the decoder LM. Yet we know very little about how these crucial MAs transform representations. Here we examine the MA output representation in three SLMs (SALMONN, Qwen2-Audio and Phi-4-Multimodal-Instruct). By finding the nearest decoder LM token to an MA representation, we uncover two strategies …

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[1/4]:
- Intelligence Primer
Karl Fezer, Andrew Sloss
https://arxiv.or…

Crosslisted article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[4/5]:
- RepoDebug: Repository-Level Multi-Task and Multi-Language Debugging Evaluation of Large Language ...
Liu, Liu, Cheng, He, Shi, Guo, Zhu, Guo, Wang, Wang

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[6/6]:
- VarCoNet: A variability-aware self-supervised framework for functional connectome extraction from...
Charalampos Lamprou, Aamna Alshehhi, Leontios J. Hadjileontiadis, Mohamed L. Seghier

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[4/6]:
- JointDiT: Enhancing RGB-Depth Joint Modeling with Diffusion Transformers
Kwon Byung-Ki, Qi Dai, Lee Hyoseok, Chong Luo, Tae-Hyun Oh

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[5/6]:
- Pre-training Limited Memory Language Models with Internal and External Knowledge
Zhao, Zalouk, Belardi, Lovelace, Zhou, Noonan, Go, Weinberger, Artzi, Sun

Can Large Language Models Bridge the Gap in Environmental Knowledge?
Linda Smail (College of Interdisciplinary Studies, Zayed University, UAE), David Santandreu Calonge (Department of Academic Development, Mohamed bin Zayed University of Artificial Intelligence, UAE), Firuz Kamalov (School of Engineering, Applied Science,Technology, Canadian University Dubai, UAE), Nur H. Orak (Department of Environmental Engineering, Marmara University, T\"urkiye)

Can Large Language Models Bridge the Gap in Environmental Knowledge?
This research investigates the potential of Artificial Intelligence (AI) models to bridge the knowledge gap in environmental education among university students. By focusing on prominent large language models (LLMs) such as GPT-3.5, GPT-4, GPT-4o, Gemini, Claude Sonnet, and Llama 2, the study assesses their effectiveness in conveying environmental concepts and, consequently, facilitating environmental education. The investigation employs a standardized tool, the Environmental Knowledge Test (EK…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[3/6]:
- Grounding the Ungrounded: A Spectral-Graph Framework for Quantifying Hallucinations in multimodal...
Supratik Sarkar, Swagatam Das

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[2/6]:
- OT Score: An OT based Confidence Score for Source Free Unsupervised Domain Adaptation
Yiming Zhang, Sitong Liu, Alex Cloninger

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/6]:
- The Challenges of Hyperparameter Tuning for Accurate Causal Effect Estimation
Damian Machlanski, Spyridon Samothrakis, Paul Clarke

Low-probability Tokens Sustain Exploration in Reinforcement Learning with Verifiable Reward
Guanhua Huang, Tingqiang Xu, Mingze Wang, Qi Yi, Xue Gong, Siheng Li, Ruibin Xiong, Kejiao Li, Yuhao Jiang, Bo Zhou
https://arxiv.org/abs/2510.03222

Low-probability Tokens Sustain Exploration in Reinforcement Learning with Verifiable Reward
Reinforcement Learning with Verifiable Rewards (RLVR) has propelled Large Language Models in complex reasoning, yet its scalability is often hindered by a training bottleneck where performance plateaus as policy entropy collapses, signaling a loss of exploration. Previous methods typically address this by maintaining high policy entropy, yet the precise mechanisms that govern meaningful exploration have remained underexplored. Our analysis suggests that an unselective focus on entropy risks amp…

To Distill or Decide? Understanding the Algorithmic Trade-off in Partially Observable Reinforcement Learning
Yuda Song, Dhruv Rohatgi, Aarti Singh, J. Andrew Bagnell
https://arxiv.org/abs/2510.03207

To Distill or Decide? Understanding the Algorithmic Trade-off in Partially Observable Reinforcement Learning
Partial observability is a notorious challenge in reinforcement learning (RL), due to the need to learn complex, history-dependent policies. Recent empirical successes have used privileged expert distillation--which leverages availability of latent state information during training (e.g., from a simulator) to learn and imitate the optimal latent, Markovian policy--to disentangle the task of "learning to see" from "learning to act". While expert distillation is more computationally efficient tha…

Best-of-Majority: Minimax-Optimal Strategy for Pass@$k$ Inference Scaling
Qiwei Di, Kaixuan Ji, Xuheng Li, Heyang Zhao, Quanquan Gu
https://arxiv.org/abs/2510.03199 https://

Best-of-Majority: Minimax-Optimal Strategy for Pass@$k$ Inference Scaling
LLM inference often generates a batch of candidates for a prompt and selects one via strategies like majority voting or Best-of- N (BoN). For difficult tasks, this single-shot selection often underperforms. Consequently, evaluations commonly report Pass@$k$: the agent may submit up to $k$ responses, and only the best of them is used when computing regret. Motivated by this, we study inference scaling in the more general Pass@$k$ inference setting, and prove that neither majority voting nor BoN …

Estimation of Resistance Training RPE using Inertial Sensors and Electromyography
James Thomas, Johan Walhstr\"om
https://arxiv.org/abs/2510.03197 https://

Estimation of Resistance Training RPE using Inertial Sensors and Electromyography
Accurate estimation of rating of perceived exertion (RPE) can enhance resistance training through personalized feedback and injury prevention. This study investigates the application of machine learning models to estimate RPE during single-arm dumbbell bicep curls, using data from wearable inertial and electromyography (EMG) sensors. A custom dataset of 69 sets and over 1000 repetitions was collected, with statistical features extracted for model training. Among the models evaluated, a random f…

Superposition disentanglement of neural representations reveals hidden alignment
Andr\'e Longon, David Klindt, Meenakshi Khosla
https://arxiv.org/abs/2510.03186 https://

Superposition disentanglement of neural representations reveals hidden alignment
The superposition hypothesis states that a single neuron within a population may participate in the representation of multiple features in order for the population to represent more features than the number of neurons. In neuroscience and AI, representational alignment metrics measure the extent to which different deep neural networks (DNNs) or brains represent similar information. In this work, we explore a critical question: \textit{does superposition interact with alignment metrics in any un…

PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning
Wanjia Zhao, Qinwei Ma, Jingzhe Shi, Shirley Wu, Jiaqi Han, Yijia Xiao, Si-Yuan Chen, Xiao Luo, Ludwig Schmidt, James Zou
https://arxiv.org/abs/2510.03185

PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning
Benchmarks for competition-style reasoning have advanced evaluation in mathematics and programming, yet physics remains comparatively explored. Most existing physics benchmarks evaluate only final answers, which fail to capture reasoning processes, while recent stepwise methods rely on heuristic LLM-as-judge scoring or restrictive linear assumptions, limiting reliability and diagnostic validity. We introduce PRISM-Physics, a process-level evaluation framework and benchmark for complex physics r…

Q-Learning with Shift-Aware Upper Confidence Bound in Non-Stationary Reinforcement Learning
Ha Manh Bui, Felix Parker, Kimia Ghobadi, Anqi Liu
https://arxiv.org/abs/2510.03181 h…

Q-Learning with Shift-Aware Upper Confidence Bound in Non-Stationary Reinforcement Learning
We study the Non-Stationary Reinforcement Learning (RL) under distribution shifts in both finite-horizon episodic and infinite-horizon discounted Markov Decision Processes (MDPs). In the finite-horizon case, the transition functions may suddenly change at a particular episode. In the infinite-horizon setting, such changes can occur at an arbitrary time step during the agent's interaction with the environment. While the Q-learning Upper Confidence Bound algorithm (QUCB) can discover a proper pol…

FTTE: Federated Learning on Resource-Constrained Devices
Irene Tenison, Anna Murphy, Charles Beauville, Lalana Kagal
https://arxiv.org/abs/2510.03165 https://

FTTE: Federated Learning on Resource-Constrained Devices
Federated learning (FL) enables collaborative model training across distributed devices while preserving data privacy, but deployment on resource-constrained edge nodes remains challenging due to limited memory, energy, and communication bandwidth. Traditional synchronous and asynchronous FL approaches further suffer from straggler induced delays and slow convergence in heterogeneous, large scale networks. We present FTTE (Federated Tiny Training Engine),a novel semi-asynchronous FL framework t…

Why Do We Need Warm-up? A Theoretical Perspective
Foivos Alimisis, Rustem Islamov, Aurelien Lucchi
https://arxiv.org/abs/2510.03164 https://arxiv.org/pdf/2…

Why Do We Need Warm-up? A Theoretical Perspective
Learning rate warm-up - increasing the learning rate at the beginning of training - has become a ubiquitous heuristic in modern deep learning, yet its theoretical foundations remain poorly understood. In this work, we provide a principled explanation for why warm-up improves training. We rely on a generalization of the $(L_0, L_1)$-smoothness condition, which bounds local curvature as a linear function of the loss sub-optimality and exhibits desirable closure properties. We demonstrate both the…

Calibrated Uncertainty Sampling for Active Learning
Ha Manh Bui, Iliana Maifeld-Carucci, Anqi Liu
https://arxiv.org/abs/2510.03162 https://arxiv.org/pdf/25…

Calibrated Uncertainty Sampling for Active Learning
We study the problem of actively learning a classifier with a low calibration error. One of the most popular Acquisition Functions (AFs) in pool-based Active Learning (AL) is querying by the model's uncertainty. However, we recognize that an uncalibrated uncertainty model on the unlabeled pool may significantly affect the AF effectiveness, leading to sub-optimal generalization and high calibration error on unseen data. Deep Neural Networks (DNNs) make it even worse as the model uncertainty from…

Mixture of Many Zero-Compute Experts: A High-Rate Quantization Theory Perspective
Yehuda Dar
https://arxiv.org/abs/2510.03151 https://arxiv.org/pdf/2510.03…

Mixture of Many Zero-Compute Experts: A High-Rate Quantization Theory Perspective
This paper uses classical high-rate quantization theory to provide new insights into mixture-of-experts (MoE) models for regression tasks. Our MoE is defined by a segmentation of the input space to regions, each with a single-parameter expert that acts as a constant predictor with zero-compute at inference. Motivated by high-rate quantization theory assumptions, we assume that the number of experts is sufficiently large to make their input-space regions very small. This lets us to study the app…

Taming Imperfect Process Verifiers: A Sampling Perspective on Backtracking
Dhruv Rohatgi, Abhishek Shetty, Donya Saless, Yuchen Li, Ankur Moitra, Andrej Risteski, Dylan J. Foster
https://arxiv.org/abs/2510.03149

Taming Imperfect Process Verifiers: A Sampling Perspective on Backtracking
Test-time algorithms that combine the generative power of language models with process verifiers that assess the quality of partial generations offer a promising lever for eliciting new reasoning capabilities, but the algorithmic design space and computational scaling properties of such approaches are still opaque, and their benefits are far from apparent when one accounts for the cost of learning a high-quality verifier. Our starting point is the observation that seemingly benign errors in a l…

Enhancing XAI Narratives through Multi-Narrative Refinement and Knowledge Distillation
Flavio Giorgi, Matteo Silvestri, Cesare Campagnano, Fabrizio Silvestri, Gabriele Tolomei
https://arxiv.org/abs/2510.03134

Enhancing XAI Narratives through Multi-Narrative Refinement and Knowledge Distillation
Explainable Artificial Intelligence has become a crucial area of research, aiming to demystify the decision-making processes of deep learning models. Among various explainability techniques, counterfactual explanations have been proven particularly promising, as they offer insights into model behavior by highlighting minimal changes that would alter a prediction. Despite their potential, these explanations are often complex and technical, making them difficult for non-experts to interpret. To a…

Signature-Informed Transformer for Asset Allocation
Yoontae Hwang, Stefan Zohren
https://arxiv.org/abs/2510.03129 https://arxiv.org/pdf/2510.03129

Signature-Informed Transformer for Asset Allocation
Robust asset allocation is a key challenge in quantitative finance, where deep-learning forecasters often fail due to objective mismatch and error amplification. We introduce the Signature-Informed Transformer (SIT), a novel framework that learns end-to-end allocation policies by directly optimizing a risk-aware financial objective. SIT's core innovations include path signatures for a rich geometric representation of asset dynamics and a signature-augmented attention mechanism embedding financi…

Real Time Headway Predictions in Urban Rail Systems and Implications for Service Control: A Deep Learning Approach
Muhammad Usama, Haris Koutsopoulos
https://arxiv.org/abs/2510.03121

Real Time Headway Predictions in Urban Rail Systems and Implications for Service Control: A Deep Learning Approach
Efficient real-time dispatching in urban metro systems is essential for ensuring service reliability, maximizing resource utilization, and improving passenger satisfaction. This study presents a novel deep learning framework centered on a Convolutional Long Short-Term Memory (ConvLSTM) model designed to predict the complex spatiotemporal propagation of train headways across an entire metro line. By directly incorporating planned terminal headways as a critical input alongside historical headway…

AdaBet: Gradient-free Layer Selection for Efficient Training of Deep Neural Networks
Irene Tenison, Soumyajit Chatterjee, Fahim Kawsar, Mohammad Malekzadeh
https://arxiv.org/abs/2510.03101

AdaBet: Gradient-free Layer Selection for Efficient Training of Deep Neural Networks
To utilize pre-trained neural networks on edge and mobile devices, we often require efficient adaptation to user-specific runtime data distributions while operating under limited compute and memory resources. On-device retraining with a target dataset can facilitate such adaptations; however, it remains impractical due to the increasing depth of modern neural nets, as well as the computational overhead associated with gradient-based optimization across all layers. Current approaches reduce trai…

Adaptive Node Feature Selection For Graph Neural Networks
Ali Azizpour, Madeline Navarro, Santiago Segarra
https://arxiv.org/abs/2510.03096 https://arxiv.o…

Adaptive Node Feature Selection For Graph Neural Networks
We propose an adaptive node feature selection approach for graph neural networks (GNNs) that identifies and removes unnecessary features during training. The ability to measure how features contribute to model output is key for interpreting decisions, reducing dimensionality, and even improving performance by eliminating unhelpful variables. However, graph-structured data introduces complex dependencies that may not be amenable to classical feature importance metrics. Inspired by this challenge…

Distilled Protein Backbone Generation
Liyang Xie, Haoran Zhang, Zhendong Wang, Wesley Tansey, Mingyuan Zhou
https://arxiv.org/abs/2510.03095 https://arxiv.…

Distilled Protein Backbone Generation
Diffusion- and flow-based generative models have recently demonstrated strong performance in protein backbone generation tasks, offering unprecedented capabilities for de novo protein design. However, while achieving notable performance in generation quality, these models are limited by their generating speed, often requiring hundreds of iterative steps in the reverse-diffusion process. This computational bottleneck limits their practical utility in large-scale protein discovery, where thousand…

Bootstrap Learning for Combinatorial Graph Alignment with Sequential GNNs
Marc Lelarge
https://arxiv.org/abs/2510.03086 https://arxiv.org/pdf/2510.03086

Bootstrap Learning for Combinatorial Graph Alignment with Sequential GNNs
Graph neural networks (GNNs) have struggled to outperform traditional optimization methods on combinatorial problems, limiting their practical impact. We address this gap by introducing a novel chaining procedure for the graph alignment problem, a fundamental NP-hard task of finding optimal node correspondences between unlabeled graphs using only structural information. Our method trains a sequence of GNNs where each network learns to iteratively refine similarity matrices produced by previous …

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem
Mingfeng Fan, Jiaqi Cheng, Yaoxin Wu, Yifeng Zhang, Yibin Yang, Guohua Wu, Guillaume Sartoretti
https://arxiv.org/abs/2510.03065

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem
In recent years, deep reinforcement learning (DRL) has gained traction for solving the NP-hard traveling salesman problem (TSP). However, limited attention has been given to the close-enough TSP (CETSP), primarily due to the challenge introduced by its neighborhood-based visitation criterion, wherein a node is considered visited if the agent enters a compact neighborhood around it. In this work, we formulate a Markov decision process (MDP) for CETSP using a discretization scheme and propose a n…

Comparative Analysis of Parameterized Action Actor-Critic Reinforcement Learning Algorithms for Web Search Match Plan Generation
Ubayd Bapoo, Clement N Nyirenda
https://arxiv.org/abs/2510.03064

Comparative Analysis of Parameterized Action Actor-Critic Reinforcement Learning Algorithms for Web Search Match Plan Generation
This study evaluates the performance of Soft Actor Critic (SAC), Greedy Actor Critic (GAC), and Truncated Quantile Critics (TQC) in high-dimensional decision-making tasks using fully observable environments. The focus is on parametrized action (PA) spaces, eliminating the need for recurrent networks, with benchmarks Platform-v0 and Goal-v0 testing discrete actions linked to continuous action-parameter spaces. Hyperparameter optimization was performed with Microsoft NNI, ensuring reproducibility…

ZeroShotOpt: Towards Zero-Shot Pretrained Models for Efficient Black-Box Optimization
Jamison Meindl, Yunsheng Tian, Tony Cui, Veronika Thost, Zhang-Wei Hong, Johannes D\"urholt, Jie Chen, Wojciech Matusik, Mina Konakovi\'c Lukovi\'c
https://arxiv.org/abs/2510.03051

ZeroShotOpt: Towards Zero-Shot Pretrained Models for Efficient Black-Box Optimization
Global optimization of expensive, derivative-free black-box functions requires extreme sample efficiency. While Bayesian optimization (BO) is the current state-of-the-art, its performance hinges on surrogate and acquisition function hyper-parameters that are often hand-tuned and fail to generalize across problem landscapes. We present ZeroShotOpt, a general-purpose, pretrained model for continuous black-box optimization tasks ranging from 2D to 20D. Our approach leverages offline reinforcement …

Bayesian E(3)-Equivariant Interatomic Potential with Iterative Restratification of Many-body Message Passing
Soohaeng Yoo Willow, Tae Hyeon Park, Gi Beom Sim, Sung Wook Moon, Seung Kyu Min, D. ChangMo Yang, Hyun Woo Kim, Juho Lee, Chang Woo Myung
https://arxiv.org/abs/2510.03046

Bayesian E(3)-Equivariant Interatomic Potential with Iterative Restratification of Many-body Message Passing
Machine learning potentials (MLPs) have become essential for large-scale atomistic simulations, enabling ab initio-level accuracy with computational efficiency. However, current MLPs struggle with uncertainty quantification, limiting their reliability for active learning, calibration, and out-of-distribution (OOD) detection. We address these challenges by developing Bayesian E(3) equivariant MLPs with iterative restratification of many-body message passing. Our approach introduces the joint ene…

CHORD: Customizing Hybrid-precision On-device Model for Sequential Recommendation with Device-cloud Collaboration
Tianqi Liu, Kairui Fu, Shengyu Zhang, Wenyan Fan, Zhaocheng Du, Jieming Zhu, Fan Wu, Fei Wu
https://arxiv.org/abs/2510.03038

CHORD: Customizing Hybrid-precision On-device Model for Sequential Recommendation with Device-cloud Collaboration
With the advancement of mobile device capabilities, deploying reranking models directly on devices has become feasible, enabling real-time contextual recommendations. When migrating models from cloud to devices, resource heterogeneity inevitably necessitates model compression. Recent quantization methods show promise for efficient deployment, yet they overlook device-specific user interests, resulting in compromised recommendation accuracy. While on-device finetuning captures personalized user …

Lightweight Transformer for EEG Classification via Balanced Signed Graph Algorithm Unrolling
Junyi Yao, Parham Eftekhar, Gene Cheung, Xujin Chris Liu, Yao Wang, Wei Hu
https://arxiv.org/abs/2510.03027 …

Lightweight Transformer for EEG Classification via Balanced Signed Graph Algorithm Unrolling
Samples of brain signals collected by EEG sensors have inherent anti-correlations that are well modeled by negative edges in a finite graph. To differentiate epilepsy patients from healthy subjects using collected EEG signals, we build lightweight and interpretable transformer-like neural nets by unrolling a spectral denoising algorithm for signals on a balanced signed graph -- graph with no cycles of odd number of negative edges. A balanced signed graph has well-defined frequencies that map to…

Differentially Private Wasserstein Barycenters
Anming Gu, Sasidhar Kunapuli, Mark Bun, Edward Chien, Kristjan Greenewald
https://arxiv.org/abs/2510.03021 https://

Differentially Private Wasserstein Barycenters
The Wasserstein barycenter is defined as the mean of a set of probability measures under the optimal transport metric, and has numerous applications spanning machine learning, statistics, and computer graphics. In practice these input measures are empirical distributions built from sensitive datasets, motivating a differentially private (DP) treatment. We present, to our knowledge, the first algorithms for computing Wasserstein barycenters under differential privacy. Empirically, on synthetic d…

Learning Robust Diffusion Models from Imprecise Supervision
Dong-Dong Wu, Jiacheng Cui, Wei Wang, Zhiqiang She, Masashi Sugiyama
https://arxiv.org/abs/2510.03016 https://…

Learning Robust Diffusion Models from Imprecise Supervision
Conditional diffusion models have achieved remarkable success in various generative tasks recently, but their training typically relies on large-scale datasets that inevitably contain imprecise information in conditional inputs. Such supervision, often stemming from noisy, ambiguous, or incomplete labels, will cause condition mismatch and degrade generation quality. To address this challenge, we propose DMIS, a unified framework for training robust Diffusion Models from Imprecise Supervision, w…

Distributional Inverse Reinforcement Learning
Feiyang Wu, Ye Zhao, Anqi Wu
https://arxiv.org/abs/2510.03013 https://arxiv.org/pdf/2510.03013

Distributional Inverse Reinforcement Learning
We propose a distributional framework for offline Inverse Reinforcement Learning (IRL) that jointly models uncertainty over reward functions and full distributions of returns. Unlike conventional IRL approaches that recover a deterministic reward estimate or match only expected returns, our method captures richer structure in expert behavior, particularly in learning the reward distribution, by minimizing first-order stochastic dominance (FSD) violations and thus integrating distortion risk mea…

BrainIB : Leveraging Graph Neural Networks and Information Bottleneck for Functional Brain Biomarkers in Schizophrenia
Tianzheng Hu, Qiang Li, Shu Liu, Vince D. Calhoun, Guido van Wingen, Shujian Yu
https://arxiv.org/abs/2510.03004

BrainIB++: Leveraging Graph Neural Networks and Information Bottleneck for Functional Brain Biomarkers in Schizophrenia
The development of diagnostic models is gaining traction in the field of psychiatric disorders. Recently, machine learning classifiers based on resting-state functional magnetic resonance imaging (rs-fMRI) have been developed to identify brain biomarkers that differentiate psychiatric disorders from healthy controls. However, conventional machine learning-based diagnostic models often depend on extensive feature engineering, which introduces bias through manual intervention. While deep learning…

From high-frequency sensors to noon reports: Using transfer learning for shaft power prediction in maritime
Akriti Sharma, Dogan Altan, Dusica Marijan, Arnbj{\o}rn Maressa
https://arxiv.org/abs/2510.03003

From high-frequency sensors to noon reports: Using transfer learning for shaft power prediction in maritime
With the growth of global maritime transportation, energy optimization has become crucial for reducing costs and ensuring operational efficiency. Shaft power is the mechanical power transmitted from the engine to the shaft and directly impacts fuel consumption, making its accurate prediction a paramount step in optimizing vessel performance. Power consumption is highly correlated with ship parameters such as speed and shaft rotation per minute, as well as weather and sea conditions. Frequent ac…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[2/5]:
- Imputation-free Learning of Tabular Data with Missing Values using Incremental Feature Partitions...
Manar D. Samad, Kazi Fuad B. Akhter, Shourav B. Rabbani, Ibna Kowsar

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[5/5]:
- Generalized and Unified Equivalences between Hardness and Pseudoentropy
Lunjia Hu, Salil Vadhan

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[3/5]:
- Towards Reasoning for PDE Foundation Models: A Reward-Model-Driven Inference-Time-Scaling Algorithm
Siddharth Mansingh, et al.

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/5]:
- Reservoir kernels and Volterra series
Lukas Gonon, Lyudmila Grigoryeva, Juan-Pablo Ortega

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[3/3]:
- Unobtrusive In-Situ Measurement of Behavior Change by Deep Metric Similarity Learning of Motion P...
Merz, Schach, Fiedler, Lugrin, Wienrich, Latoschik

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[2/3]:
- Testing for correlation between network structure and high-dimensional node covariates
Alexander Fuchs-Kreiss, Keith Levin

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/3]:
- A Small Dataset May Go a Long Way: Process Duration Prediction in Clinical Settings
Harald St\"orrle, Anastasia Hort

ChronoGraph: A Real-World Graph-Based Multivariate Time Series Dataset
Adrian Catalin Lutu, Ioana Pintilie, Elena Burceanu, Andrei Manolache
https://arxiv.org/abs/2509.04449 htt…

ChronoGraph: A Real-World Graph-Based Multivariate Time Series Dataset
We present ChronoGraph, a graph-structured multivariate time series forecasting dataset built from real-world production microservices. Each node is a service that emits a multivariate stream of system-level performance metrics, capturing CPU, memory, and network usage patterns, while directed edges encode dependencies between services. The primary task is forecasting future values of these signals at the service level. In addition, ChronoGraph provides expert-annotated incident windows as anom…

Towards Cognitively-Faithful Decision-Making Models to Improve AI Alignment
Cyrus Cousins, Vijay Keswani, Vincent Conitzer, Hoda Heidari, Jana Schaich Borg, Walter Sinnott-Armstrong
https://arxiv.org/abs/2509.04445

Towards Cognitively-Faithful Decision-Making Models to Improve AI Alignment
Recent AI work trends towards incorporating human-centric objectives, with the explicit goal of aligning AI models to personal preferences and societal values. Using standard preference elicitation methods, researchers and practitioners build models of human decisions and judgments, which are then used to align AI behavior with that of humans. However, models commonly used in such elicitation processes often do not capture the true cognitive processes of human decision making, such as when peop…

Delta Activations: A Representation for Finetuned Large Language Models
Zhiqiu Xu, Amish Sethi, Mayur Naik, Ser-Nam Lim
https://arxiv.org/abs/2509.04442 https://

Delta Activations: A Representation for Finetuned Large Language Models
The success of powerful open source Large Language Models (LLMs) has enabled the community to create a vast collection of post-trained models adapted to specific tasks and domains. However, navigating and understanding these models remains challenging due to inconsistent metadata and unstructured repositories. We introduce Delta Activations, a method to represent finetuned models as vector embeddings by measuring shifts in their internal activations relative to a base model. This representation…

Unveiling the Role of Data Uncertainty in Tabular Deep Learning
Nikolay Kartashev, Ivan Rubachev, Artem Babenko
https://arxiv.org/abs/2509.04430 https://ar…

Unveiling the Role of Data Uncertainty in Tabular Deep Learning
Recent advancements in tabular deep learning have demonstrated exceptional practical performance, yet the field often lacks a clear understanding of why these techniques actually succeed. To address this gap, our paper highlights the importance of the concept of data uncertainty for explaining the effectiveness of the recent tabular DL methods. In particular, we reveal that the success of many beneficial design choices in tabular DL, such as numerical feature embeddings, retrieval-augmented mod…

Echo State Networks as State-Space Models: A Systems Perspective
Pradeep Singh, Balasubramanian Raman
https://arxiv.org/abs/2509.04422 https://arxiv.org/pd…

Echo State Networks as State-Space Models: A Systems Perspective
Echo State Networks (ESNs) are typically presented as efficient, readout-trained recurrent models, yet their dynamics and design are often guided by heuristics rather than first principles. We recast ESNs explicitly as state-space models (SSMs), providing a unified systems-theoretic account that links reservoir computing with classical identification and modern kernelized SSMs. First, we show that the echo-state property is an instance of input-to-state stability for a contractive nonlinear SSM…

Towards a Unified View of Large Language Model Post-Training
Xingtai Lv, Yuxin Zuo, Youbang Sun, Hongyi Liu, Yuntian Wei, Zhekai Chen, Lixuan He, Xuekai Zhu, Kaiyan Zhang, Bingning Wang, Ning Ding, Bowen Zhou
https://arxiv.org/abs/2509.04419

Towards a Unified View of Large Language Model Post-Training
Two major sources of training data exist for post-training modern language models: online (model-generated rollouts) data, and offline (human or other-model demonstrations) data. These two types of data are typically used by approaches like Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT), respectively. In this paper, we show that these approaches are not in contradiction, but are instances of a single optimization process. We derive a Unified Policy Gradient Estimator, and present …

Interpretable Clustering with Adaptive Heterogeneous Causal Structure Learning in Mixed Observational Data
Wenrui Li, Qinghao Zhang, Xiaowo Wang
https://arxiv.org/abs/2509.04415

Interpretable Clustering with Adaptive Heterogeneous Causal Structure Learning in Mixed Observational Data
Understanding causal heterogeneity is essential for scientific discovery in domains such as biology and medicine. However, existing methods lack causal awareness, with insufficient modeling of heterogeneity, confounding, and observational constraints, leading to poor interpretability and difficulty distinguishing true causal heterogeneity from spurious associations. We propose an unsupervised framework, HCL (Interpretable Causal Mechanism-Aware Clustering with Adaptive Heterogeneous Causal Stru…

IPA: An Information-Preserving Input Projection Framework for Efficient Foundation Model Adaptation
Yuan Yin, Shashanka Venkataramanan, Tuan-Hung Vu, Andrei Bursuc, Matthieu Cord
https://arxiv.org/abs/2509.04398

IPA: An Information-Preserving Input Projection Framework for Efficient Foundation Model Adaptation
Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, reduce adaptation cost by injecting low-rank updates into pretrained weights. However, LoRA's down-projection is randomly initialized and data-agnostic, discarding potentially useful information. Prior analyses show that this projection changes little during training, while the up-projection carries most of the adaptation, making the random input compression a performance bottleneck. We propose IPA, a feature-aware projection framewo…

Transition Models: Rethinking the Generative Learning Objective
Zidong Wang, Yiyuan Zhang, Xiaoyu Yue, Xiangyu Yue, Yangguang Li, Wanli Ouyang, Lei Bai
https://arxiv.org/abs/2509.04394

Transition Models: Rethinking the Generative Learning Objective
A fundamental dilemma in generative modeling persists: iterative diffusion models achieve outstanding fidelity, but at a significant computational cost, while efficient few-step alternatives are constrained by a hard quality ceiling. This conflict between generation steps and output quality arises from restrictive training objectives that focus exclusively on either infinitesimal dynamics (PF-ODEs) or direct endpoint prediction. We address this challenge by introducing an exact, continuous-time…