Tootfinder

Kyle Shanahan and 49ers Players Address the Media Following Joint Practices https://www.49ers.com/video/kyle-shanahan-and-49ers-players-address-the-media-following-joint-practices

lkml_reply: Linux kernel reply network
A network of replies among email address found on the Linux kernel mailing list. Nodes are email addresses and each directed edge represents a reply from email address i to email address j. The date of this snapshot is uncertain.
This network has 63399 nodes and 1096440 edges.
Tags: Social, Communication, Unweighted, Multigraph, Timestamps

Still blows my mind that in Germany (and some other European countries) it’s mandatory to put an imprint with your actual address on your personal website. So basically doxing by default, unless you pay a provider to hide behind a proxy address to forward legal letters.*
* There are exemption for private, non-commercial sites but that boils down to .htaccess protection. Anything else is not considered private or non-commercial in legal practice.

'No idea what you have unleashed': Charlie Kirk's wife delivers first public address (USA Today)
https://www.usatoday.com/story/news/nation/2025/09/12/charlie-kirks-wife-delivers-first-public-address/86123453007/
http://www.memeorandum.com/250913/p13#a250913p13

If you send a bulk mail to a retired email address (former staff member, for example) and your unsubscribe feedback page doesn't have "address no longer used", I *will* click the "this is spam" feedback item.

We've Issued Our First IP Address Certificate
#LetsEncrypt

ParallelSearch: Train your LLMs to Decompose Query and Search Sub-queries in Parallel with Reinforcement Learning
Shu Zhao, Tan Yu, Anbang Xu, Japinder Singh, Aaditya Shukla, Rama Akkiraju
https://arxiv.org/abs/2508.09303

ParallelSearch: Train your LLMs to Decompose Query and Search Sub-queries in Parallel with Reinforcement Learning
Reasoning-augmented search agents such as Search-R1, trained via reinforcement learning with verifiable rewards (RLVR), demonstrate remarkable capabilities in multi-step information retrieval from external knowledge sources. These agents address the limitations of their parametric memory by dynamically gathering relevant facts to address complex reasoning tasks. However, existing approaches suffer from a fundamental architectural limitation: they process search queries strictly sequentially, ev…

rCamInspector: Building Reliability and Trust on IoT (Spy) Camera Detection using XAI
Priyanka Rushikesh Chaudhary, Manan Gupta, Jabez Christopher, Putrevu Venkata Sai Charan, Rajib Ranjan Maiti
https://arxiv.org/abs/2509.09989

rCamInspector: Building Reliability and Trust on IoT (Spy) Camera Detection using XAI
The classification of network traffic using machine learning (ML) models is one of the primary mechanisms to address the security issues in IoT networks and/or IoT devices. However, the ML models often act as black-boxes that create a roadblock to take critical decision based on the model output. To address this problem, we design and develop a system, called rCamInspector, that employs Explainable AI (XAI) to provide reliable and trustworthy explanations to model output. rCamInspector adopts t…

MLM: Learning Multi-task Loco-Manipulation Whole-Body Control for Quadruped Robot with Arm
Xin Liu, Bida Ma, Chenkun Qi, Yan Ding, Zhaxizhuoma, Guorong Zhang, Pengan Chen, Kehui Liu, Zhongjie Jia, Chuyue Guan, Yule Mo, Jiaqi Liu, Feng Gao, Jiangwei Zhong, Bin Zhao, Xuelong Li
https://arxiv.org/abs/2508.10538…

MLM: Learning Multi-task Loco-Manipulation Whole-Body Control for Quadruped Robot with Arm
Whole-body loco-manipulation for quadruped robots with arm remains a challenging problem, particularly in achieving multi-task control. To address this, we propose MLM, a reinforcement learning framework driven by both real-world and simulation data. It enables a six-DoF robotic arm--equipped quadruped robot to perform whole-body loco-manipulation for multiple tasks autonomously or under human teleoperation. To address the problem of balancing multiple tasks during the learning of loco-manipula…

Maybe I'll fire up a disposable email address and reply, see how much I can talk the scammer up to before I get bored.

This is not a joke... A press release from the UK government about a meeting of the "national drought group", in order to "save water at home", suggests to:
"Delete old emails and pictures as data centres require vast amounts of water to cool their systems"
... while the government also pus…

National Drought Group meets to address “nationally significant” water shortfall
Five areas remain in drought with six more in prolonged dry weather

“The Maha report is essentially a distraction from the real causes of poor health,”
said Elizabeth Jacobs, professor emerita at the University of Arizona and a founding member of Defend Public Health.
“This administration does not want to address things like poverty and education and access to healthcare.
Instead, they’re distracting the public with information on solutions to problems that don’t actually exist.
When the foundation of your policy is not evidence-based…

EPA Grants Were Set to Address Health Risks on the Hopi and Navajo Reservations, Until the Trump Administration Cut Them - Inside Climate News
https://insideclimatenews.org/news/15062025/cut-epa-grants-hopi-navajo-reservations/

Dynamical dark energy in models close to $\Lambda$CDM
Saikat Chakraborty, Charlotte Louw, Peter K. S. Dunsby, Kelly MacDevette, Alvaro de la Cruz Dombriz
https://arxiv.org/abs/2508.09813

Dynamical dark energy in models close to $Λ$CDM
In this communication we address whether or not there is an equivalence between the kinematical and dynamical descriptions of the spatially flat $Λ$CDM model. We address this by investigating whether an almost $Λ$CDM expansion history ($j(z)\approx1$) corresponds to an almost $Λ$CDM model ($w_{\rm DE}(z)\approx-1$) by considering two particular explicit examples. At least for the cases considered, this turns out not to be the case. Instead, what we find is that an almost $Λ$CDM cosmic evolu…

@… For example, we were trying to figure out why Thing A could not connect to Thing B. Turns out, we hadn’t allowed the IP range. This is on us. However, we had to jump through 4871 hoops to figure this out, because we have to express IP address ranges, not intent, and therefore the cloud UI cannot help us figure out why there’s a discrepancy.
It doesn’t help that…

"Global wetlands conference results in resolutions for protection & restoration"
#Wetlands #Environment

Global wetlands conference results in resolutions for restoration & protection
Wetlands are disappearing at three times the rate of forests, alarming conservationists who highlight the ecosystems’ role in achieving global biodiversity and climate goals. To address these challenges and chart a path toward sustainable conservation, delegates gathered in Zimbabwe for the 15th Meeting of the Conference of the Contracting Parties (COP15) on Wetlands in July. […]

Nun dampfe ich 18 Seiten Text auf 10 ein, erhöhe den Salbaderanteil dabei zugleich und verkaufe das Ganze als "Presidential Address".

so is #wordpress discriminating against Italians?
(to be fair, you can log in even if you don't tick the first box)
#pizzahawaii

Anyone who follows me has probably gathered that I've become a fan of Discourse after running a few Discourse instances. They are currently requesting feedback.
I was pleased to see that they used a survey vendor that appears to have meaningful a11y — I didn't look for aria labels so I won't speak to that, but what I did notice is that there is explicitly-labeled support for keyboard-only interaction on every page, and I really appreciated that thoughtfulness.

#FreeTube be like "#YouTube has blocked your IP address from watching videos. Please try switching to a different VPN or proxy." 😔

Maybe if the Democrats actually engaged with the electorate in ALL districts to understand and address concerns they wouldn't think they needed to resort to the idiotic Redistricting plan Newsom proposes .
But, then, Newsom got where he is by catering to his base, not seriously addressing issues.
https:…

Maybe if the Democrats actually engaged with the electorate in ALL districts to understand and address concerns they wouldn't think they needed to resort to the idiotic Redistricting plan Newsom proposes .
But, then, Newsom got where he is by catering to his base, not seriously addressing issues.
https:…

For those folks who used to use the old vinylrecords(at)a.gup.pe account, that account is now dead, along with all of the gup.pe accounts, so be sure to unfollow it and follow @vinylrecords(at-symbol)fedigroups.social instead. Tag your vinyl posts with that address to have your post boosted to all followers.
Please boost this to get the word out!

“FOR REVIEW: Address formats around the world”
https://www.w3.org/blog/International/2025/08/06/for-review-address-formats-around-the-world/
The doc to review:

{\mu}Fork: Supporting POSIX fork Within a Single-Address-Space OS
John Alistair Kressel, Hugo Lefeuvre, Pierre Olivier
https://arxiv.org/abs/2509.09439 https://

μFork: Supporting POSIX fork Within a Single-Address-Space OS
Single-address-space operating systems have well-known lightweightness benefits that result from their central design idea: the kernel and applications share a unique address space. This model makes these operating systems (OSes) incompatible by design with a large class of software: multiprocess POSIX applications. Indeed, the semantics of the primitive used to create POSIX processes, fork, are inextricably tied to the existence of multiple address spaces. Prior approaches addressing this is…

Obesity & diet
I wouldn't normally share a positive story about the new diet drugs, because I've seen someone get obsessed with them who was at a perfectly acceptable weight *by majority standards* (surprise: every weight is in fact perfectly acceptable by *objective* standards, because every "weight-associated" health risk is its own danger that should be assessed *in individuals*). I think two almost-contradictory things:
1. In a society shuddering under the burden of metastasized fatmisia, there's a very real danger in promoting the new diet drugs because lots of people who really don't need them will be psychologically bullied into using them and suffer from the cost and/or side effects.
2. For many individuals under the assault of our society's fatmisia, "just ignore it" is not a sufficient response, and also for specific people for whom decreasing their weight can address *specific* health risks/conditions that they *want* to address that way, these drugs can be a useful tool.
I know @… to be a trustworthy & considerate person, so I think it's responsible to share this:
#Fat #Diet #Obesity

I am hereby doxxing my own address

Well this is fun, #Ohio HB96 seems that it would require me to pay to to collect and store your state issued ID if you want to browse Mastodon via this instance.
That's fine, I don't mind spending money which I don't have to collect users' private data including your home address and full legal name. You're fine with giving effectively a complete stranger acro…

Exploring Expert Specialization through Unsupervised Training in Sparse Mixture of Experts
Strahinja Nikolic, Ilker Oguz, Demetri Psaltis
https://arxiv.org/abs/2509.10025 https:…

Exploring Expert Specialization through Unsupervised Training in Sparse Mixture of Experts
Understanding the internal organization of neural networks remains a fundamental challenge in deep learning interpretability. We address this challenge by exploring a novel Sparse Mixture of Experts Variational Autoencoder (SMoE-VAE) architecture. We test our model on the QuickDraw dataset, comparing unsupervised expert routing against a supervised baseline guided by ground-truth labels. Surprisingly, we find that unsupervised routing consistently achieves superior reconstruction performance. T…

Self-Supervised Stereo Matching with Multi-Baseline Contrastive Learning
Peng Xu, Zhiyu Xiang, Jingyun Fu, Tianyu Pu, Kai Wang, Chaojie Ji, Tingming Bai, Eryun Liu
https://arxiv.org/abs/2508.10838

Self-Supervised Stereo Matching with Multi-Baseline Contrastive Learning
Current self-supervised stereo matching relies on the photometric consistency assumption, which breaks down in occluded regions due to ill-posed correspondences. To address this issue, we propose BaCon-Stereo, a simple yet effective contrastive learning framework for self-supervised stereo network training in both non-occluded and occluded regions. We adopt a teacher-student paradigm with multi-baseline inputs, in which the stereo pairs fed into the teacher and student share the same reference …

Improving and Evaluating Open Deep Research Agents
Doaa Allabadi, Kyle Bradbury, Jordan M. Malof
https://arxiv.org/abs/2508.10152 https://arxiv.org/pdf/250…

Improving and Evaluating Open Deep Research Agents
We focus here on Deep Research Agents (DRAs), which are systems that can take a natural language prompt from a user, and then autonomously search for, and utilize, internet-based content to address the prompt. Recent DRAs have demonstrated impressive capabilities on public benchmarks however, recent research largely involves proprietary closed-source systems. At the time of this work, we only found one open-source DRA, termed Open Deep Research (ODR). In this work we adapt the challenging recen…

Government that’s going all in on “AI” urges you to delete old emails and pictures to save water. You couldn’t make this shit up. https://tech.lgbt/@Natasha_Jay/115016993942039602

Natasha Jay (@Natasha_Jay@tech.lgbt)
Attached: 1 image Did AI recommend this? In a new digital twist to environmental responsibility, the #UK government is now suggesting people to “Delete old emails and pictures” in data centres to help with the current drought 🤪 This country ... (no words) https://www.gov.uk/government/news/national-drought-group-meets-to-address-nationally-significant-water-shortfall

from my link log —
Building accurate postal address matching systems.
https://www.robinlinacre.com/address_matching/
saved 2025-06-30

Sources: investigators have uncovered evidence that Russia is at least in part responsible for a recent hack of the US federal court filing system (New York Times)
https://www.nytimes.com/2025/08/12/us/politics/russia-hack-federal-court-system.html…

Russia Is Suspected to Be Behind Breach of Federal Court Filing System
Federal officials are scrambling to assess the damage and address flaws in a sprawling, heavily used computer system long known to have vulnerabilities.

Exploring Quantum Annealing for Coarse-Grained Protein Folding
Timon Scheiber, Matthias Heller, Andreas Giebel
https://arxiv.org/abs/2508.10660 https://arx…

Exploring Quantum Annealing for Coarse-Grained Protein Folding
We explore the potential application of quantum annealing to address the protein structure problem. To this end, we compare several proposed ab initio protein folding models for quantum computers and analyze their scaling and performance for classical and quantum heuristics. Furthermore, we introduce a novel encoding of coordinate based models on the tetrahedral lattice, based on interleaved grids. Our findings reveal significant variations in model performance, with one model yielding unphysic…

Hey, I've previously mentioned Michael Sweater's cartooning stuff here (https://sweater.substack.com/).
I was nice to him, so I received 3 x free 1 month subscriptions to give away. So if you're interested, DM me an email address to receive Substack posts. First in, best spammed!
KTXHBAI!

Sweater Weather | Michael Sweater | Substack
Comics and writing straight from the desk of Michael Sweater. Click to read Sweater Weather, by Michael Sweater, a Substack publication with thousands of subscribers.

email_company: Manufacturing company email (2010)
A network of emails among employee email addresses at a mid-sized manufacturing company. Each directed edge represents an email sent from address i to address j. Edges are timestamped, and occurred over a 6 month period in an unspecified year.
This network has 167 nodes and 82927 edges.
Tags: Social, Communication, Unweighted, Multigraph, Timestamps

Oh #ShadowServer, now you're just being silly...
#InfoSec #Scanners

CLEAR: Calibrated Learning for Epistemic and Aleatoric Risk
Ilia Azizi, Juraj Bodik, Jakob Heiss, Bin Yu
https://arxiv.org/abs/2507.08150 https://

CLEAR: Calibrated Learning for Epistemic and Aleatoric Risk
Accurate uncertainty quantification is critical for reliable predictive modeling, especially in regression tasks. Existing methods typically address either aleatoric uncertainty from measurement noise or epistemic uncertainty from limited data, but not necessarily both in a balanced way. We propose CLEAR, a calibration method with two distinct parameters, $γ_1$ and $γ_2$, to combine the two uncertainty components for improved conditional coverage. CLEAR is compatible with any pair of aleatori…

Maybe it's just me, but if my husband died two days ago the last thing on my mind would be 'vowing to continue the movement he founded'
‘The movement my husband built will not die’: Charlie Kirk’s widow vows to continue slain activist’s work | The Independent
https://www.independent.co.uk/news/world/americas/us-politics/charlie-kirk-widow-erika-first-address-b2825772.html

An interesting guide to what you can do for climate action:
https://bsky.app/profile/globalecoguy.bsky.social/post/3lysibvowyk26

Dr. Jonathan Foley (@globalecoguy.bsky.social)
We often get asked, "What can I *personally* do to help address climate change?" Sadly, there hasn't been a very good answer to this question. Previous guides often treated people as monoliths, providing a one-size-fits-all answer. But that's all changed, thanks to SHIFT. jointheshift.earth https://jointheshift.earth

An interesting guide to what you can do for climate action:
https://bsky.app/profile/globalecoguy.bsky.social/post/3lysibvowyk26

Dr. Jonathan Foley (@globalecoguy.bsky.social)
We often get asked, "What can I *personally* do to help address climate change?" Sadly, there hasn't been a very good answer to this question. Previous guides often treated people as monoliths, providing a one-size-fits-all answer. But that's all changed, thanks to SHIFT. jointheshift.earth https://jointheshift.earth

Russia Is Suspected to Be Behind Breach of Federal Court Filing System
Federal officials are scrambling to assess the damage and address flaws in a sprawling, heavily used computer system long known to have vulnerabilities.
Russia is at least in part responsible for a recent hack of the computer system that manages federal court documents,
including highly sensitive records that might contain information
that could reveal sources and people charged with national secur…

Russia Is Suspected to Be Behind Breach of Federal Court Filing System
Federal officials are scrambling to assess the damage and address flaws in a sprawling, heavily used computer system long known to have vulnerabilities.

Arabic Large Language Models for Medical Text Generation
Abdulrahman Allam, Seif Ahmed, Ali Hamdi, Ammar Mohammed
https://arxiv.org/abs/2509.10095 https://…

Arabic Large Language Models for Medical Text Generation
Efficient hospital management systems (HMS) are critical worldwide to address challenges such as overcrowding, limited resources, and poor availability of urgent health care. Existing methods often lack the ability to provide accurate, real-time medical advice, particularly for irregular inputs and underrepresented languages. To overcome these limitations, this study proposes an approach that fine-tunes large language models (LLMs) for Arabic medical text generation. The system is designed to a…

January Food Benchmark (JFB): A Public Benchmark Dataset and Evaluation Suite for Multimodal Food Analysis
Amir Hosseinian, Ashkan Dehghani Zahedani, Umer Mansoor, Noosheen Hashemi, Mark Woodward
https://arxiv.org/abs/2508.09966

January Food Benchmark (JFB): A Public Benchmark Dataset and Evaluation Suite for Multimodal Food Analysis
Progress in AI for automated nutritional analysis is critically hampered by the lack of standardized evaluation methodologies and high-quality, real-world benchmark datasets. To address this, we introduce three primary contributions. First, we present the January Food Benchmark (JFB), a publicly available collection of 1,000 food images with human-validated annotations. Second, we detail a comprehensive benchmarking framework, including robust metrics and a novel, application-oriented overall s…

A Markovian Framing of WaveFunctionCollapse for Procedurally Generating Aesthetically Complex Environments
Franklin Yiu, Mohan Lu, Nina Li, Kevin Joseph, Tianxu Zhang, Julian Togelius, Timothy Merino, Sam Earle
https://arxiv.org/abs/2509.09919

A Markovian Framing of WaveFunctionCollapse for Procedurally Generating Aesthetically Complex Environments
Procedural content generation often requires satisfying both designer-specified objectives and adjacency constraints implicitly imposed by the underlying tile set. To address the challenges of jointly optimizing both constraints and objectives, we reformulate WaveFunctionCollapse (WFC) as a Markov Decision Process (MDP), enabling external optimization algorithms to focus exclusively on objective maximization while leveraging WFC's propagation mechanism to enforce constraint satisfaction. We emp…

MDNS: Masked Diffusion Neural Sampler via Stochastic Optimal Control
Yuchen Zhu, Wei Guo, Jaemoo Choi, Guan-Horng Liu, Yongxin Chen, Molei Tao
https://arxiv.org/abs/2508.10684 h…

MDNS: Masked Diffusion Neural Sampler via Stochastic Optimal Control
We study the problem of learning a neural sampler to generate samples from discrete state spaces where the target probability mass function $π\propto\mathrm{e}^{-U}$ is known up to a normalizing constant, which is an important task in fields such as statistical physics, machine learning, combinatorial optimization, etc. To better address this challenging task when the state space has a large cardinality and the distribution is multi-modal, we propose $\textbf{M}$asked $\textbf{D}$iffusion $\te…

BERTector: Intrusion Detection Based on Joint-Dataset Learning
Haoyang Hu, Xun Huang, Chenyu Wu, Shiwen Liu, Zhichao Lian, Shuangquan Zhang
https://arxiv.org/abs/2508.10327 http…

BERTector: Intrusion Detection Based on Joint-Dataset Learning
Intrusion detection systems (IDS) are facing challenges in generalization and robustness due to the heterogeneity of network traffic and the diversity of attack patterns. To address this issue, we propose a new joint-dataset training paradigm for IDS and propose a scalable BERTector framework based on BERT. BERTector integrates three key components: NSS-Tokenizer for traffic-aware semantic tokenization, supervised fine-tuning with a hybrid dataset, and low-rank adaptation (LoRA) for efficient t…

LiDAR, GNSS and IMU Sensor Alignment through Dynamic Time Warping to Construct 3D City Maps
Haitian Wang, Hezam Albaqami, Xinyu Wang, Muhammad Ibrahim, Zainy M. Malakan, Abdullah M. Algamdi, Mohammed H. Alghamdi, Ajmal Mian
https://arxiv.org/abs/2507.08420

LiDAR, GNSS and IMU Sensor Alignment through Dynamic Time Warping to Construct 3D City Maps
LiDAR-based 3D mapping suffers from cumulative drift causing global misalignment, particularly in GNSS-constrained environments. To address this, we propose a unified framework that fuses LiDAR, GNSS, and IMU data for high-resolution city-scale mapping. The method performs velocity-based temporal alignment using Dynamic Time Warping and refines GNSS and IMU signals via extended Kalman filtering. Local maps are built using Normal Distributions Transform-based registration and pose graph optimiza…

Routing and Wavelength Assignment with Minimal Attack Radius for QKD Networks
Mengyao Li, Qiaolun Zhang, Zongshuai Yang, Stefano Bregni, Alberto Gatto, Raouf Boutaba, Massimo Tornatore
https://arxiv.org/abs/2508.10613

Routing and Wavelength Assignment with Minimal Attack Radius for QKD Networks
Quantum Key Distribution (QKD) can distribute keys with guaranteed security but remains susceptible to key exchange interruption due to physical-layer threats, such as high-power jamming attacks. To address this challenge, we first introduce a novel metric, namely Maximum Number of Affected Requests (maxNAR), to quantify the worst-case impact of a single physical-layer attack, and then we investigate a new problem of Routing and Wavelength Assignment with Minimal Attack Radius (RWA-MAR). We for…

LLMC : Benchmarking Vision-Language Model Compression with a Plug-and-play Toolkit
Chengtao Lv, Bilang Zhang, Yang Yong, Ruihao Gong, Yushi Huang, Shiqiao Gu, Jiajun Wu, Yumeng Shi, Jinyang Guo, Wenya Wang
https://arxiv.org/abs/2508.09981

LLMC+: Benchmarking Vision-Language Model Compression with a Plug-and-play Toolkit
Large Vision-Language Models (VLMs) exhibit impressive multi-modal capabilities but suffer from prohibitive computational and memory demands, due to their long visual token sequences and massive parameter sizes. To address these issues, recent works have proposed training-free compression methods. However, existing efforts often suffer from three major limitations: (1) Current approaches do not decompose techniques into comparable modules, hindering fair evaluation across spatial and temporal r…

UDA: Unsupervised Debiasing Alignment for Pair-wise LLM-as-a-Judge
Yang Zhang, Cunxiang Wang, Lindong Wu, Wenbo Yu, Yidong Wang, Guangsheng Bao, Jie Tang
https://arxiv.org/abs/2508.09724

UDA: Unsupervised Debiasing Alignment for Pair-wise LLM-as-a-Judge
Pairwise evaluation of Large Language Models (LLMs) is a common paradigm, but it is prone to preference bias, where judges systematically favor certain outputs, such as their own. This bias leads to inconsistent and skewed rankings across different judges. To address this, we first empirically demonstrate significant and heterogeneous biases in cross-model evaluations. We then propose UDA (Unsupervised Debiasing Alignment), a framework that reduces inter-judge disagreement by dynamically adjust…

Finding Common Ground: Using Large Language Models to Detect Agreement in Multi-Agent Decision Conferences
Selina Heller, Mohamed Ibrahim, David Antony Selby, Sebastian Vollmer
https://arxiv.org/abs/2507.08440

Finding Common Ground: Using Large Language Models to Detect Agreement in Multi-Agent Decision Conferences
Decision conferences are structured, collaborative meetings that bring together experts from various fields to address complex issues and reach a consensus on recommendations for future actions or policies. These conferences often rely on facilitated discussions to ensure productive dialogue and collective agreement. Recently, Large Language Models (LLMs) have shown significant promise in simulating real-world scenarios, particularly through collaborative multi-agent systems that mimic group in…

APFL: Analytic Personalized Federated Learning via Dual-Stream Least Squares
Kejia Fan, Jianheng Tang, Zhirui Yang, Feijiang Han, Jiaxu Li, Run He, Yajiang Huang, Anfeng Liu, Houbing Herbert Song, Yunhuai Liu, Huiping Zhuang
https://arxiv.org/abs/2508.10732

APFL: Analytic Personalized Federated Learning via Dual-Stream Least Squares
Personalized Federated Learning (PFL) has presented a significant challenge to deliver personalized models to individual clients through collaborative training. Existing PFL methods are often vulnerable to non-IID data, which severely hinders collective generalization and then compromises the subsequent personalization efforts. In this paper, to address this non-IID issue in PFL, we propose an Analytic Personalized Federated Learning (APFL) approach via dual-stream least squares. In our APFL, w…

CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model
Zhuoyuan Yu, Yuxing Long, Zihan Yang, Chengyan Zeng, Hongwei Fan, Jiyao Zhang, Hao Dong
https://arxiv.org/abs/2508.10416

CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model
Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a me…

A Context-aware Attention and Graph Neural Network-based Multimodal Framework for Misogyny Detection
Mohammad Zia Ur Rehman, Sufyaan Zahoor, Areeb Manzoor, Musharaf Maqbool, Nagendra Kumar
https://arxiv.org/abs/2508.09175

A Context-aware Attention and Graph Neural Network-based Multimodal Framework for Misogyny Detection
A substantial portion of offensive content on social media is directed towards women. Since the approaches for general offensive content detection face a challenge in detecting misogynistic content, it requires solutions tailored to address offensive content against women. To this end, we propose a novel multimodal framework for the detection of misogynistic and sexist content. The framework comprises three modules: the Multimodal Attention module (MANM), the Graph-based Feature Reconstruction …

LeanRAG: Knowledge-Graph-Based Generation with Semantic Aggregation and Hierarchical Retrieval
Yaoze Zhang, Rong Wu, Pinlong Cai, Xiaoman Wang, Guohang Yan, Song Mao, Ding Wang, Botian Shi
https://arxiv.org/abs/2508.10391

LeanRAG: Knowledge-Graph-Based Generation with Semantic Aggregation and Hierarchical Retrieval
Retrieval-Augmented Generation (RAG) plays a crucial role in grounding Large Language Models by leveraging external knowledge, whereas the effectiveness is often compromised by the retrieval of contextually flawed or incomplete information. To address this, knowledge graph-based RAG methods have evolved towards hierarchical structures, organizing knowledge into multi-level summaries. However, these approaches still suffer from two critical, unaddressed challenges: high-level conceptual summarie…

Understanding Textual Emotion Through Emoji Prediction
Ethan Gordon, Nishank Kuppa, Rigved Tummala, Sriram Anasuri
https://arxiv.org/abs/2508.10222 https://

Understanding Textual Emotion Through Emoji Prediction
This project explores emoji prediction from short text sequences using four deep learning architectures: a feed-forward network, CNN, transformer, and BERT. Using the TweetEval dataset, we address class imbalance through focal loss and regularization techniques. Results show BERT achieves the highest overall performance due to its pre-training advantage, while CNN demonstrates superior efficacy on rare emoji classes. This research shows the importance of architecture selection and hyperparamete…

Stabilizing Long-term Multi-turn Reinforcement Learning with Gated Rewards
Zetian Sun, Dongfang Li, Zhuoen Chen, Yuhuai Qin, Baotian Hu
https://arxiv.org/abs/2508.10548 https://…

Stabilizing Long-term Multi-turn Reinforcement Learning with Gated Rewards
Reward sparsity in long-horizon reinforcement learning (RL) tasks remains a significant challenge, while existing outcome-based reward shaping struggles to define meaningful immediate rewards without introducing bias or requiring explicit task decomposition. Alternatively, verification-based reward shaping uses stepwise critics, but misalignment between immediate rewards and long-term objectives can lead to reward hacking and suboptimal policies. In this work, we address this problem in the con…

Prespecified-Performance Kinematic Tracking Control for Aerial Manipulation
Hauzi Cao, Jiahao Shen, Zhengzhen Li, Qinquan Ren, Shiyu Zhao
https://arxiv.org/abs/2509.10065 https:…

Prespecified-Performance Kinematic Tracking Control for Aerial Manipulation
This paper studies the kinematic tracking control problem for aerial manipulators. Existing kinematic tracking control methods, which typically employ proportional-derivative feedback or tracking-error-based feedback strategies, may fail to achieve tracking objectives within specified time constraints. To address this limitation, we propose a novel control framework comprising two key components: end-effector tracking control based on a user-defined preset trajectory and quadratic programming-b…

Story2Board: A Training-Free Approach for Expressive Storyboard Generation
David Dinkevich, Matan Levy, Omri Avrahami, Dvir Samuel, Dani Lischinski
https://arxiv.org/abs/2508.09983

Story2Board: A Training-Free Approach for Expressive Storyboard Generation
We present Story2Board, a training-free framework for expressive storyboard generation from natural language. Existing methods narrowly focus on subject identity, overlooking key aspects of visual storytelling such as spatial composition, background evolution, and narrative pacing. To address this, we introduce a lightweight consistency framework composed of two components: Latent Panel Anchoring, which preserves a shared character reference across panels, and Reciprocal Attention Value Mixing,…

MSRS: Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models
Xinyan Jiang, Lin Zhang, Jiayi Zhang, Qingsong Yang, Guimin Hu, Di Wang, Lijie Hu
https://arxiv.org/abs/2508.10599

MSRS: Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models
Activation steering offers a promising approach to controlling the behavior of Large Language Models by directly manipulating their internal activations. However, most existing methods struggle to jointly steer multiple attributes, often resulting in interference and undesirable trade-offs. To address this challenge, we propose Multi-Subspace Representation Steering (MSRS), a novel framework for effective multi-attribute steering via subspace representation fine-tuning. MSRS reduces inter-attri…

Specialised or Generic? Tokenization Choices for Radiology Language Models
Hermione Warr, Wentian Xu, Harry Anthony, Yasin Ibrahim, Daniel McGowan, Konstantinos Kamnitsas
https://arxiv.org/abs/2508.09952

Specialised or Generic? Tokenization Choices for Radiology Language Models
The vocabulary used by language models (LM) - defined by the tokenizer - plays a key role in text generation quality. However, its impact remains under-explored in radiology. In this work, we address this gap by systematically comparing general, medical, and domain-specific tokenizers on the task of radiology report summarisation across three imaging modalities. We also investigate scenarios with and without LM pre-training on PubMed abstracts. Our findings demonstrate that medical and domain-s…

REFN: A Reinforcement-Learning-From-Network Framework against 1-day/n-day Exploitations
Tianlong Yu, Lihong Liu, Ziyi Zhou, Fudu Xing, Kailong Wang, Yang Yang
https://arxiv.org/abs/2508.10701

REFN: A Reinforcement-Learning-From-Network Framework against 1-day/n-day Exploitations
The exploitation of 1 day or n day vulnerabilities poses severe threats to networked devices due to massive deployment scales and delayed patching (average Mean Time To Patch exceeds 60 days). Existing defenses, including host based patching and network based filtering, are inadequate due to limited scalability across diverse devices, compatibility issues especially with embedded or legacy systems, and error prone deployment process (manual patch validation). To address these issues, we introdu…

DECAMP: Towards Scene-Consistent Multi-Agent Motion Prediction with Disentangled Context-Aware Pre-Training
Jianxin Shi, Zengqi Peng, Xiaolong Chen, Tianyu Wo, Jun Ma
https://arxiv.org/abs/2509.10426

DECAMP: Towards Scene-Consistent Multi-Agent Motion Prediction with Disentangled Context-Aware Pre-Training
Trajectory prediction is a critical component of autonomous driving, essential for ensuring both safety and efficiency on the road. However, traditional approaches often struggle with the scarcity of labeled data and exhibit suboptimal performance in multi-agent prediction scenarios. To address these challenges, we introduce a disentangled context-aware pre-training framework for multi-agent motion prediction, named DECAMP. Unlike existing methods that entangle representation learning with pret…

Beyond conventional vision: RGB-event fusion for robust object detection in dynamic traffic scenarios
Zhanwen Liu, Yujing Sun, Yang Wang, Nan Yang, Shengbo Eben Li, Xiangmo Zhao
https://arxiv.org/abs/2508.10704

Beyond conventional vision: RGB-event fusion for robust object detection in dynamic traffic scenarios
The dynamic range limitation of conventional RGB cameras reduces global contrast and causes loss of high-frequency details such as textures and edges in complex traffic environments (e.g., nighttime driving, tunnels), hindering discriminative feature extraction and degrading frame-based object detection. To address this, we integrate a bio-inspired event camera with an RGB camera to provide high dynamic range information and propose a motion cue fusion network (MCFNet), which achieves optimal s…

RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA
Bhavik Agarwal, Hemant Sunil Jomraj, Simone Kaplunov, Jack Krolick, Viktoria Rojkova
https://arxiv.org/abs/2508.09893

RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA
Regulatory compliance question answering (QA) requires precise, verifiable information, and domain-specific expertise, posing challenges for Large Language Models (LLMs). In this work, we present a novel multi-agent framework that integrates a Knowledge Graph (KG) of Regulatory triplets with Retrieval-Augmented Generation (RAG) to address these demands. First, agents build and maintain an ontology-free KG by extracting subject--predicate--object (SPO) triplets from regulatory documents and syst…

ILT-Iterative LoRA Training through Focus-Feedback-Fix for Multilingual Speech Recognition
Qingliang Meng, Hao Wu, Wei Liang, Wei Xu, Qing Zhao
https://arxiv.org/abs/2507.08477

ILT-Iterative LoRA Training through Focus-Feedback-Fix for Multilingual Speech Recognition
The deep integration of large language models and automatic speech recognition systems has become a promising research direction with high practical value. To address the overfitting issue commonly observed in Low-Rank Adaptation (LoRA) during the supervised fine-tuning (SFT) stage, this work proposes an innovative training paradigm Iterative LoRA Training (ILT) in combination with an Iterative Pseudo Labeling strategy, effectively enhancing the theoretical upper bound of model performance. Bas…

MIMo grows! Simulating body and sensory development in a multimodal infant model
Francisco M. L\'opez, Miles Lenz, Marco G. Fedozzi, Arthur Aubret, Jochen Triesch
https://arxiv.org/abs/2509.09805

MIMo grows! Simulating body and sensory development in a multimodal infant model
Infancy is characterized by rapid body growth and an explosive change of sensory and motor abilities. However, developmental robots and simulation platforms are typically designed in the image of a specific age, which limits their ability to capture the changing abilities and constraints of developing infants. To address this issue, we present MIMo v2, a new version of the multimodal infant model. It includes a growing body with increasing actuation strength covering the age range from birth to…

IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection
Yanhui Li, Yunkang Cao, Chengliang Liu, Yuan Xiong, Xinghui Dong, Chao Huang
https://arxiv.org/abs/2508.09178

IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection
Industrial anomaly detection is a critical component of modern manufacturing, yet the scarcity of defective samples restricts traditional detection methods to scenario-specific applications. Although Vision-Language Models (VLMs) demonstrate significant advantages in generalization capabilities, their performance in industrial anomaly detection remains limited. To address this challenge, we propose IAD-R1, a universal post-training framework applicable to VLMs of different architectures and par…

The Morality of Probability: How Implicit Moral Biases in LLMs May Shape the Future of Human-AI Symbiosis
Eoin O'Doherty, Nicole Weinrauch, Andrew Talone, Uri Klempner, Xiaoyuan Yi, Xing Xie, Yi Zeng
https://arxiv.org/abs/2509.10297

The Morality of Probability: How Implicit Moral Biases in LLMs May Shape the Future of Human-AI Symbiosis
Artificial intelligence (AI) is advancing at a pace that raises urgent questions about how to align machine decision-making with human moral values. This working paper investigates how leading AI systems prioritize moral outcomes and what this reveals about the prospects for human-AI symbiosis. We address two central questions: (1) What moral values do state-of-the-art large language models (LLMs) implicitly favour when confronted with dilemmas? (2) How do differences in model architecture, cul…

SI-FACT: Mitigating Knowledge Conflict via Self-Improving Faithfulness-Aware Contrastive Tuning
Shengqiang Fu
https://arxiv.org/abs/2509.10208 https://arxi…

SI-FACT: Mitigating Knowledge Conflict via Self-Improving Faithfulness-Aware Contrastive Tuning
Large Language Models often generate unfaithful responses in knowledge intensive tasks due to knowledge conflict,that is,a preference for relying on internal parametric knowledge rather than the provided context.To address this issue,we propose a novel self improving framework,Self Improving Faithfulness Aware Contrastive Tuning.The framework uses a self instruct mechanism that allows the base LLM to automatically generate high quality,structured contrastive learning data,including anchor sampl…

What to Ask Next? Probing the Imaginative Reasoning of LLMs with TurtleSoup Puzzles
Mengtao Zhou, Sifan Wu, Huan Zhang, Qi Sima, Bang Liu
https://arxiv.org/abs/2508.10358 https:…

What to Ask Next? Probing the Imaginative Reasoning of LLMs with TurtleSoup Puzzles
We investigate the capacity of Large Language Models (LLMs) for imaginative reasoning--the proactive construction, testing, and revision of hypotheses in information-sparse environments. Existing benchmarks, often static or focused on social deduction, fail to capture the dynamic, exploratory nature of this reasoning process. To address this gap, we introduce a comprehensive research framework based on the classic "Turtle Soup" game, integrating a benchmark, an agent, and an evaluation protocol…

AEGIS: Authenticity Evaluation Benchmark for AI-Generated Video Sequences
Jieyu Li, Xin Zhang, Joey Tianyi Zhou
https://arxiv.org/abs/2508.10771 https://ar…

AEGIS: Authenticity Evaluation Benchmark for AI-Generated Video Sequences
Recent advances in AI-generated content have fueled the rise of highly realistic synthetic videos, posing severe risks to societal trust and digital integrity. Existing benchmarks for video authenticity detection typically suffer from limited realism, insufficient scale, and inadequate complexity, failing to effectively evaluate modern vision-language models against sophisticated forgeries. To address this critical gap, we introduce AEGIS, a novel large-scale benchmark explicitly targeting the …

PakBBQ: A Culturally Adapted Bias Benchmark for QA
Abdullah Hashmat, Muhammad Arham Mirza, Agha Ali Raza
https://arxiv.org/abs/2508.10186 https://arxiv.org…

PakBBQ: A Culturally Adapted Bias Benchmark for QA
With the widespread adoption of Large Language Models (LLMs) across various applications, it is empirical to ensure their fairness across all user communities. However, most LLMs are trained and evaluated on Western centric data, with little attention paid to low-resource languages and regional contexts. To address this gap, we introduce PakBBQ, a culturally and regionally adapted extension of the original Bias Benchmark for Question Answering (BBQ) dataset. PakBBQ comprises over 214 templates,…

ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing
Lingen Li, Guangzhi Wang, Zhaoyang Zhang, Yaowei Li, Xiaoyu Li, Qi Dou, Jinwei Gu, Tianfan Xue, Ying Shan
https://arxiv.org/abs/2508.10881

ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing
Traditional cartoon and anime production involves keyframing, inbetweening, and colorization stages, which require intensive manual effort. Despite recent advances in AI, existing methods often handle these stages separately, leading to error accumulation and artifacts. For instance, inbetweening approaches struggle with large motions, while colorization methods require dense per-frame sketches. To address this, we introduce ToonComposer, a generative model that unifies inbetweening and coloriz…

MEML-GRPO: Heterogeneous Multi-Expert Mutual Learning for RLVR Advancement
Weitao Jia, Jinghui Lu, Haiyang Yu, Siqi Wang, Guozhi Tang, An-Lan Wang, Weijie Yin, Dingkang Yang, Yuxiang Nie, Bin Shan, Hao Feng, Irene Li, Kun Yang, Han Wang, Jingqun Tang, Teng Fu, Changhong Jin, Chao Feng, Xiaohui Lv, Can Huang
https://arxiv.org/abs/2508.09670…

MEML-GRPO: Heterogeneous Multi-Expert Mutual Learning for RLVR Advancement
Recent advances demonstrate that reinforcement learning with verifiable rewards (RLVR) significantly enhances the reasoning capabilities of large language models (LLMs). However, standard RLVR faces challenges with reward sparsity, where zero rewards from consistently incorrect candidate answers provide no learning signal, particularly in challenging tasks. To address this, we propose Multi-Expert Mutual Learning GRPO (MEML-GRPO), an innovative framework that utilizes diverse expert prompts as …

Towards Reliable and Interpretable Document Question Answering via VLMs
Alessio Chen, Simone Giovannini, Andrea Gemelli, Fabio Coppini, Simone Marinai
https://arxiv.org/abs/2509.10129

Towards Reliable and Interpretable Document Question Answering via VLMs
Vision-Language Models (VLMs) have shown strong capabilities in document understanding, particularly in identifying and extracting textual information from complex documents. Despite this, accurately localizing answers within documents remains a major challenge, limiting both interpretability and real-world applicability. To address this, we introduce \textit{DocExplainerV0}, a plug-and-play bounding-box prediction module that decouples answer generation from spatial localization. This design m…

ESSENTIAL: Episodic and Semantic Memory Integration for Video Class-Incremental Learning
Jongseo Lee, Kyungho Bae, Kyle Min, Gyeong-Moon Park, Jinwoo Choi
https://arxiv.org/abs/2508.10896

ESSENTIAL: Episodic and Semantic Memory Integration for Video Class-Incremental Learning
In this work, we tackle the problem of video classincremental learning (VCIL). Many existing VCIL methods mitigate catastrophic forgetting by rehearsal training with a few temporally dense samples stored in episodic memory, which is memory-inefficient. Alternatively, some methods store temporally sparse samples, sacrificing essential temporal information and thereby resulting in inferior performance. To address this trade-off between memory-efficiency and performance, we propose EpiSodic and SE…

Modeling Human Responses to Multimodal AI Content
Zhiqi Shen, Shaojing Fan, Danni Xu, Terence Sim, Mohan Kankanhalli
https://arxiv.org/abs/2508.10769 https://

Modeling Human Responses to Multimodal AI Content
As AI-generated content becomes widespread, so does the risk of misinformation. While prior research has primarily focused on identifying whether content is authentic, much less is known about how such content influences human perception and behavior. In domains like trading or the stock market, predicting how people react (e.g., whether a news post will go viral), can be more critical than verifying its factual accuracy. To address this, we take a human-centered approach and introduce the MhAI…

CMHG: A Dataset and Benchmark for Headline Generation of Minority Languages in China
Guixian Xu, Zeli Su, Ziyin Zhang, Jianing Liu, XU Han, Ting Zhang, Yushuang Dong
https://arxiv.org/abs/2509.09990

CMHG: A Dataset and Benchmark for Headline Generation of Minority Languages in China
Minority languages in China, such as Tibetan, Uyghur, and Traditional Mongolian, face significant challenges due to their unique writing systems, which differ from international standards. This discrepancy has led to a severe lack of relevant corpora, particularly for supervised tasks like headline generation. To address this gap, we introduce a novel dataset, Chinese Minority Headline Generation (CMHG), which includes 100,000 entries for Tibetan, and 50,000 entries each for Uyghur and Mongolia…

XAgents: A Unified Framework for Multi-Agent Cooperation via IF-THEN Rules and Multipolar Task Processing Graph
Hailong Yang, Mingxian Gu, Jianqi Wang, Guanjin Wang, Zhaohong Deng
https://arxiv.org/abs/2509.10054

XAgents: A Unified Framework for Multi-Agent Cooperation via IF-THEN Rules and Multipolar Task Processing Graph
The rapid advancement of Large Language Models (LLMs) has significantly enhanced the capabilities of Multi-Agent Systems (MAS) in supporting humans with complex, real-world tasks. However, MAS still face challenges in effective task planning when handling highly complex tasks with uncertainty, often resulting in misleading or incorrect outputs that hinder task execution. To address this, we propose XAgents, a unified multi-agent cooperative framework built on a multipolar task processing graph …

SGPMIL: Sparse Gaussian Process Multiple Instance Learning
Andreas Lolos, Stergios Christodoulidis, Maria Vakalopoulou, Jose Dolz, Aris Moustakas
https://arxiv.org/abs/2507.08711 …

SGPMIL: Sparse Gaussian Process Multiple Instance Learning
Multiple Instance Learning (MIL) offers a natural solution for settings where only coarse, bag-level labels are available, without having access to instance-level annotations. This is usually the case in digital pathology, which consists of gigapixel sized images. While deterministic attention-based MIL approaches achieve strong bag-level performance, they often overlook the uncertainty inherent in instance relevance. In this paper, we address the lack of uncertainty quantification in instance-…

DeepDive: Advancing Deep Search Agents with Knowledge Graphs and Multi-Turn RL
Rui Lu, Zhenyu Hou, Zihan Wang, Hanchen Zhang, Xiao Liu, Yujiang Li, Shi Feng, Jie Tang, Yuxiao Dong
https://arxiv.org/abs/2509.10446

DeepDive: Advancing Deep Search Agents with Knowledge Graphs and Multi-Turn RL
Augmenting large language models (LLMs) with browsing tools substantially improves their potential as deep search agents to solve complex, real-world tasks. Yet, open LLMs still perform poorly in such settings due to limited long-horizon reasoning capacity with browsing tools and the lack of sufficiently difficult supervised data. To address these challenges, we present DeepDive to advance deep search agents. First, we propose a strategy to automatically synthesize complex, difficult, and hard-…

Multi-Agent Trust Region Policy Optimisation: A Joint Constraint Approach
Chak Lam Shek, Guangyao Shi, Pratap Tokekar
https://arxiv.org/abs/2508.10340 https://

Multi-Agent Trust Region Policy Optimisation: A Joint Constraint Approach
Multi-agent reinforcement learning (MARL) requires coordinated and stable policy updates among interacting agents. Heterogeneous-Agent Trust Region Policy Optimization (HATRPO) enforces per-agent trust region constraints using Kullback-Leibler (KL) divergence to stabilize training. However, assigning each agent the same KL threshold can lead to slow and locally optimal updates, especially in heterogeneous settings. To address this limitation, we propose two approaches for allocating the KL dive…

FineState-Bench: A Comprehensive Benchmark for Fine-Grained State Control in GUI Agents
Fengxian Ji, Jingpu Yang, Zirui Song, Yuanxi Wang, Zhexuan Cui, Yuke Li, Qian Jiang, Miao Fang, Xiuying Chen
https://arxiv.org/abs/2508.09241

FineState-Bench: A Comprehensive Benchmark for Fine-Grained State Control in GUI Agents
With the rapid advancement of generative artificial intelligence technology, Graphical User Interface (GUI) agents have demonstrated tremendous potential for autonomously managing daily tasks through natural language instructions. However, current evaluation frameworks for GUI agents suffer from fundamental flaws: existing benchmarks overly focus on coarse-grained task completion while neglecting fine-grained control capabilities crucial for real-world applications. To address this, we introduc…

Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization
Itai Mondshine, Tzuf Paz-Argaman, Reut Tsarfaty
https://arxiv.org/abs/2507.08342

Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization
Automatic n-gram based metrics such as ROUGE are widely used for evaluating generative tasks such as summarization. While these metrics are considered indicative (even if imperfect) of human evaluation for English, their suitability for other languages remains unclear. To address this, we systematically assess evaluation metrics for generation both n-gram-based and neural based to evaluate their effectiveness across languages and tasks. Specifically, we design a large-scale evaluation suite acr…

PASS: Probabilistic Agentic Supernet Sampling for Interpretable and Adaptive Chest X-Ray Reasoning
Yushi Feng, Junye Du, Yingying Hong, Qifan Wang, Lequan Yu
https://arxiv.org/abs/2508.10501

PASS: Probabilistic Agentic Supernet Sampling for Interpretable and Adaptive Chest X-Ray Reasoning
Existing tool-augmented agentic systems are limited in the real world by (i) black-box reasoning steps that undermine trust of decision-making and pose safety risks, (ii) poor multimodal integration, which is inherently critical for healthcare tasks, and (iii) rigid and computationally inefficient agentic pipelines. We introduce PASS (Probabilistic Agentic Supernet Sampling), the first multimodal framework to address these challenges in the context of Chest X-Ray (CXR) reasoning. PASS adaptivel…

Reinforced Language Models for Sequential Decision Making
Jim Dilkes, Vahid Yazdanpanah, Sebastian Stein
https://arxiv.org/abs/2508.10839 https://arxiv.org…

Reinforced Language Models for Sequential Decision Making
Large Language Models (LLMs) show potential as sequential decision-making agents, but their application is often limited due to a reliance on large, computationally expensive models. This creates a need to improve smaller models, yet existing post-training methods are designed for single-turn interactions and cannot handle credit assignment in multi-step agentic tasks. To address this, we introduce Multi-Step Group-Relative Policy Optimization (MS-GRPO), a new algorithm for post-training LLM ag…

EvoCurr: Self-evolving Curriculum with Behavior Code Generation for Complex Decision-making
Yang Cheng, Zilai Wang, Weiyu Ma, Wenhui Zhu, Yue Deng, Jian Zhao
https://arxiv.org/abs/2508.09586

EvoCurr: Self-evolving Curriculum with Behavior Code Generation for Complex Decision-making
Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse domains, including programming, planning, and decision-making. However, their performance often degrades when faced with highly complex problem instances that require deep reasoning over long horizons. In such cases, direct problem-solving approaches can lead to inefficiency or failure due to the lack of structured intermediate guidance. To address this, we propose a novel self-evolve framework, EvoCurr, in wh…

From Black Box to Transparency: Enhancing Automated Interpreting Assessment with Explainable AI in College Classrooms
Zhaokun Jiang, Ziyin Zhang
https://arxiv.org/abs/2508.10860

From Black Box to Transparency: Enhancing Automated Interpreting Assessment with Explainable AI in College Classrooms
Recent advancements in machine learning have spurred growing interests in automated interpreting quality assessment. Nevertheless, existing research suffers from insufficient examination of language use quality, unsatisfactory modeling effectiveness due to data scarcity and imbalance, and a lack of efforts to explain model predictions. To address these gaps, we propose a multi-dimensional modeling framework that integrates feature engineering, data augmentation, and explainable machine learning…

The (R)evolution of Scientific Workflows in the Agentic AI Era: Towards Autonomous Science
Woong Shin, Renan Souza, Daniel Rosendo, Fr\'ed\'eric Suter, Feiyi Wang, Prasanna Balaprakash, Rafael Ferreira da Silva
https://arxiv.org/abs/2509.09915

The (R)evolution of Scientific Workflows in the Agentic AI Era: Towards Autonomous Science
Modern scientific discovery increasingly requires coordinating distributed facilities and heterogeneous resources, forcing researchers to act as manual workflow coordinators rather than scientists. Advances in AI leading to AI agents show exciting new opportunities that can accelerate scientific discovery by providing intelligence as a component in the ecosystem. However, it is unclear how this new capability would materialize and integrate in the real world. To address this, we propose a conce…

KELPS: A Framework for Verified Multi-Language Autoformalization via Semantic-Syntactic Alignment
Jiyao Zhang, Chengli Zhong, Hui Xu, Qige Li, Yi Zhou
https://arxiv.org/abs/2507.08665

KELPS: A Framework for Verified Multi-Language Autoformalization via Semantic-Syntactic Alignment
Modern large language models (LLMs) show promising progress in formalizing informal mathematics into machine-verifiable theorems. However, these methods still face bottlenecks due to the limited quantity and quality of multilingual parallel corpora. In this paper, we propose a novel neuro-symbolic framework KELPS (Knowledge-Equation based Logical Processing System) to address these problems. KELPS is an iterative framework for translating, synthesizing, and filtering informal data into multiple…

AWorld: Dynamic Multi-Agent System with Stable Maneuvering for Robust GAIA Problem Solving
Zhitian Xie, Qintong Wu, Chengyue Yu, Chenyi Zhuang, Jinjie Gu
https://arxiv.org/abs/2508.09889

AWorld: Dynamic Multi-Agent System with Stable Maneuvering for Robust GAIA Problem Solving
The rapid advancement of large language models (LLMs) has empowered intelligent agents to leverage diverse external tools for solving complex real-world problems. However, as agents increasingly depend on multiple tools, they encounter new challenges: extended contexts from disparate sources and noisy or irrelevant tool outputs can undermine system reliability and accuracy. These challenges underscore the necessity for enhanced stability in agent-based systems. To address this, we introduce dyn…

Continuous Bangla Sign Language Translation: Mitigating the Expense of Gloss Annotation with the Assistance of Graph
Safaeid Hossain Arib, Rabeya Akter, Sejuti Rahman
https://arxiv.org/abs/2508.10687

Continuous Bangla Sign Language Translation: Mitigating the Expense of Gloss Annotation with the Assistance of Graph
Millions of individuals worldwide are affected by deafness and hearing impairment. Sign language serves as a sophisticated means of communication for the deaf and hard of hearing. However, in societies that prioritize spoken languages, sign language often faces underestimation, leading to communication barriers and social exclusion. The Continuous Bangla Sign Language Translation project aims to address this gap by enhancing translation methods. While recent approaches leverage transformer arch…

Value Function Initialization for Knowledge Transfer and Jump-start in Deep Reinforcement Learning
Soumia Mehimeh
https://arxiv.org/abs/2508.09277 https://…

Value Function Initialization for Knowledge Transfer and Jump-start in Deep Reinforcement Learning
Value function initialization (VFI) is an effective way to achieve a jumpstart in reinforcement learning (RL) by leveraging value estimates from prior tasks. While this approach is well established in tabular settings, extending it to deep reinforcement learning (DRL) poses challenges due to the continuous nature of the state-action space, the noisy approximations of neural networks, and the impracticality of storing all past models for reuse. In this work, we address these challenges and intro…

Which one Performs Better? Wav2Vec or Whisper? Applying both in Badini Kurdish Speech to Text (BKSTT)
Renas Adnan, Hossein Hassani
https://arxiv.org/abs/2508.09957 https://

Which one Performs Better? Wav2Vec or Whisper? Applying both in Badini Kurdish Speech to Text (BKSTT)
Speech-to-text (STT) systems have a wide range of applications. They are available in many languages, albeit at different quality levels. Although Kurdish is considered a less-resourced language from a processing perspective, SST is available for some of the Kurdish dialects, for instance, Sorani (Central Kurdish). However, that is not applied to other Kurdish dialects, Badini and Hawrami, for example. This research is an attempt to address this gap. Bandin, approximately, has two million speak…

GAMA: A General Anonymizing Multi-Agent System for Privacy Preservation Enhanced by Domain Rules and Disproof Method
Hailong Yang, Renhuo Zhao, Guanjin Wang, Zhaohong Deng
https://arxiv.org/abs/2509.10018

GAMA: A General Anonymizing Multi-Agent System for Privacy Preservation Enhanced by Domain Rules and Disproof Method
With the rapid advancement of Large Language Model (LLM), LLM-based agents exhibit exceptional abilities in understanding and generating natural language, facilitating human-like collaboration and information transmission in LLM-based Multi-Agent System (MAS). High-performance LLMs are often hosted on remote servers in public spaces. When tasks involve privacy data, MAS cannot securely utilize these LLMs without implementing privacy-preserving mechanisms. To address this challenge, we propose a…

Benchmarking Vision-Language Models on Chinese Ancient Documents: From OCR to Knowledge Reasoning
Haiyang Yu, Yuchuan Wu, Fan Shi, Lei Liao, Jinghui Lu, Xiaodong Ge, Han Wang, Minghan Zhuo, Xuecheng Wu, Xiang Fei, Hao Feng, Guozhi Tang, An-Lan Wang, Hanshen Zhu, Yangfan He, Quanhuan Liang, Liyuan Meng, Chao Feng, Can Huang, Jingqun Tang, Bin Li
https://

Benchmarking Vision-Language Models on Chinese Ancient Documents: From OCR to Knowledge Reasoning
Chinese ancient documents, invaluable carriers of millennia of Chinese history and culture, hold rich knowledge across diverse fields but face challenges in digitization and understanding, i.e., traditional methods only scan images, while current Vision-Language Models (VLMs) struggle with their visual and linguistic complexity. Existing document benchmarks focus on English printed texts or simplified Chinese, leaving a gap for evaluating VLMs on ancient Chinese documents. To address this, we p…

KG-Attention: Knowledge Graph-Guided Attention at Test-Time via Bidirectional Information Aggregation
Songlin Zhai, Guilin Qi, Yuan Meng
https://arxiv.org/abs/2507.08704

KG-Attention: Knowledge Graph-Guided Attention at Test-Time via Bidirectional Information Aggregation
Knowledge graphs (KGs) play a critical role in enhancing large language models (LLMs) by introducing structured and grounded knowledge into the learning process. However, most existing KG-enhanced approaches rely on parameter-intensive fine-tuning, which risks catastrophic forgetting and degrades the pretrained model's generalization. Moreover, they exhibit limited adaptability to real-time knowledge updates due to their static integration frameworks. To address these issues, we introduce the f…