Tootfinder

OpenAI releases GPT-5 pro, a version with extended reasoning exclusive to ChatGPT Pro subscribers, saying it scored 88.4% without tools on the GPQA benchmark (Maximilian Schreiner/The Decoder)
https://the-decoder.com/openai-claims-

OpenAI launches GPT-5 as a unified system with adaptive reasoning for complex tasks
OpenAI has unveiled GPT-5, a new AI system that builds on the reasoning advances of the o1 and o3 models and unifies every previous model line into a single adaptive architecture.

OmniEAR: Benchmarking Agent Reasoning in Embodied Tasks
Zixuan Wang, Dingming Li, Hongxing Li, Shuo Chen, Yuchen Yan, Wenqi Zhang, Yongliang Shen, Weiming Lu, Jun Xiao, Yueting Zhuang
https://arxiv.org/abs/2508.05614

OmniEAR: Benchmarking Agent Reasoning in Embodied Tasks
Large language models excel at abstract reasoning but their capacity for embodied agent reasoning remains largely unexplored. We present OmniEAR, a comprehensive framework for evaluating how language models reason about physical interactions, tool usage, and multi-agent coordination in embodied tasks. Unlike existing benchmarks that provide predefined tool sets or explicit collaboration directives, OmniEAR requires agents to dynamically acquire capabilities and autonomously determine coordinati…

Driver Assistant: Persuading Drivers to Adjust Secondary Tasks Using Large Language Models
Wei Xiang, Muchen Li, Jie Yan, Manling Zheng, Hanfei Zhu, Mengyun Jiang, Lingyun Sun
https://arxiv.org/abs/2508.05238

Driver Assistant: Persuading Drivers to Adjust Secondary Tasks Using Large Language Models
Level 3 automated driving systems allows drivers to engage in secondary tasks while diminishing their perception of risk. In the event of an emergency necessitating driver intervention, the system will alert the driver with a limited window for reaction and imposing a substantial cognitive burden. To address this challenge, this study employs a Large Language Model (LLM) to assist drivers in maintaining an appropriate attention on road conditions through a "humanized" persuasive advice. Our too…

Experimental Analysis of Productive Interaction Strategy with ChatGPT: User Study on Function and Project-level Code Generation Tasks
Sangwon Hyun, Hyunjun Kim, Jinhyuk Jang, Hyojin Choi, M. Ali Babar
https://arxiv.org/abs/2508.04125

Experimental Analysis of Productive Interaction Strategy with ChatGPT: User Study on Function and Project-level Code Generation Tasks
The application of Large Language Models (LLMs) is growing in the productive completion of Software Engineering tasks. Yet, studies investigating the productive prompting techniques often employed a limited problem space, primarily focusing on well-known prompting patterns and mainly targeting function-level SE practices. We identify significant gaps in real-world workflows that involve complexities beyond class-level (e.g., multi-class dependencies) and different features that can impact Human…

Real-Time Iteration Scheme for Diffusion Policy
Yufei Duan, Hang Yin, Danica Kragic
https://arxiv.org/abs/2508.05396 https://arxiv.org/pdf/2508.05396

Real-Time Iteration Scheme for Diffusion Policy
Diffusion Policies have demonstrated impressive performance in robotic manipulation tasks. However, their long inference time, resulting from an extensive iterative denoising process, and the need to execute an action chunk before the next prediction to maintain consistent actions limit their applicability to latency-critical tasks or simple tasks with a short cycle time. While recent methods explored distillation or alternative policy structures to accelerate inference, these often demand addi…

MateInfoUB: A Real-World Benchmark for Testing LLMs in Competitive, Multilingual, and Multimodal Educational Tasks
Dumitran Adrian Marius, Theodor-Pierre Moroianu, Buca Mihnea-Vicentiu
https://arxiv.org/abs/2507.03162

MateInfoUB: A Real-World Benchmark for Testing LLMs in Competitive, Multilingual, and Multimodal Educational Tasks
The rapid advancement of Large Language Models (LLMs) has transformed various domains, particularly computer science (CS) education. These models exhibit remarkable capabilities in code-related tasks and problem-solving, raising questions about their potential and limitations in advanced CS contexts. This study presents a novel bilingual (English-Romanian) multimodal (text and image) dataset of multiple-choice questions derived from a high-level computer science competition. A particularity of …

BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks
Sagnik Anupam, Davis Brown, Shuo Li, Eric Wong, Hamed Hassani, Osbert Bastani
https://arxiv.org/abs/2510.02418

BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks
LLM web agents now browse and take actions on the open web, yet current agent evaluations are constrained to sandboxed environments or artificial tasks. We introduce BrowserArena, a live open-web agent evaluation platform that collects user-submitted tasks, runs Arena-style head-to-head comparisons, and uses step-level human feedback to surface failure modes. Collecting and analyzing step-level annotations on the agent traces, we identify three consistent failure modes: captcha resolution, pop-…

Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Vision
Luozheng Qin, Jia Gong, Yuqing Sun, Tianjiao Li, Mengping Yang, Xiaomeng Yang, Chao Qu, Zhiyu Tan, Hao Li
https://arxiv.org/abs/2508.05606

Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Vision
Chain-of-Thought (CoT) reasoning has been widely adopted to enhance Large Language Models (LLMs) by decomposing complex tasks into simpler, sequential subtasks. However, extending CoT to vision-language reasoning tasks remains challenging, as it often requires interpreting transitions of visual states to support reasoning. Existing methods often struggle with this due to limited capacity of modeling visual state transitions or incoherent visual trajectories caused by fragmented architectures. …

Todays tasks:
Migrating my personal matrix-server
Hopefully migrate my writefreely blogs
Do something stupid or sleep early

FurtherAI, which uses AI to automate insurance tasks such as claims processing, raised a $25M Series A led by a16z, bringing its total funding to $30M (Chris Metinko/Axios)
https://www.axios.com/pro/enterprise-software-deals/2025/10…

Exclusive: FurtherAI raises $25 million to automate insurance tasks
The a16z-led Series A comes as the insurance industry is moving fast to adopt AI.

windsurfers: Windsurfers network (1986)
A network of interpersonal contacts among windsurfers in southern California during the Fall of 1986. The edge weights indicate the perception of social affiliations majored by the tasks in which each individual was asked​ to sort cards with other surfer’s name in the order of closeness.
This network has 43 nodes and 336 edges.
Tags: Social, Offline, Weighted

SAFERad: A Framework to Enable Radar Data for Safety-Relevant Perception Tasks
Tim Br\"uhl, Jenny Gl\"onkler, Robin Schwager, Tin Stribor Sohn, Tim Dieter Eberhardt, S\"oren Hohmann
https://arxiv.org/abs/2507.03959

SAFERad: A Framework to Enable Radar Data for Safety-Relevant Perception Tasks
Radar sensors play a crucial role for perception systems in automated driving but suffer from a high level of noise. In the past, this could be solved by strict filters, which remove most false positives at the expense of undetected objects. Future highly automated functions are much more demanding with respect to error rate. Hence, if the radar sensor serves as a component of perception systems for such functions, a simple filter strategy cannot be applied. In this paper, we present a modified…

Learning Linear Regression with Low-Rank Tasks in-Context
Kaito Takanami, Takashi Takahashi, Yoshiyuki Kabashima
https://arxiv.org/abs/2510.04548 https://a…

Learning Linear Regression with Low-Rank Tasks in-Context
In-context learning (ICL) is a key building block of modern large language models, yet its theoretical mechanisms remain poorly understood. It is particularly mysterious how ICL operates in real-world applications where tasks have a common structure. In this work, we address this problem by analyzing a linear attention model trained on low-rank regression tasks. Within this setting, we precisely characterize the distribution of predictions and the generalization error in the high-dimensional li…

Latency Minimization for Multi-AAV-Enabled ISCC Systems with Movable Antenna
Yiyang Chen, Wenchao Liu, Chunjie Wang, Yinyu Wu, Xuhui Zhang, Yanyan Shen
https://arxiv.org/abs/2508.05574

Latency Minimization for Multi-AAV-Enabled ISCC Systems with Movable Antenna
This paper investigates an autonomous aerial vehicle (AAV)-enabled integrated sensing, communication, and computation system, with a particular focus on integrating movable antennas (MAs) into the system for enhancing overall system performance. Specifically, multiple MA-enabled AVVs perform sensing tasks and simultaneously transmit the generated computational tasks to the base station for processing. To minimize the maximum latency under the sensing and resource constraints, we formulate an op…

When Should Users Check? A Decision-Theoretic Model of Confirmation Frequency in Multi-Step AI Agent Tasks
Jieyu Zhou, Aryan Roy, Sneh Gupta, Daniel Weitekamp, Christopher J. MacLellan
https://arxiv.org/abs/2510.05307

When Should Users Check? A Decision-Theoretic Model of Confirmation Frequency in Multi-Step AI Agent Tasks
Existing AI agents typically execute multi-step tasks autonomously and only allow user confirmation at the end. During execution, users have little control, making the confirm-at-end approach brittle: a single error can cascade and force a complete restart. Confirming every step avoids such failures, but imposes tedious overhead. Balancing excessive interruptions against costly rollbacks remains an open challenge. We address this problem by modeling confirmation as a minimum time scheduling pro…

Foundation models for equation discovery in high energy physics
Manuel Morales-Alvarado
https://arxiv.org/abs/2510.03397 https://arxiv.org/pdf/2510.03397…

Foundation models for equation discovery in high energy physics
Foundation models, large machine learning models trained on broad, multimodal datasets, have been gaining increasing attention in scientific applications due to their strong performance on diverse downstream tasks. Large Language Models (LLMs), a prominent instance of foundation models, have achieved remarkable success in tasks such as text and image generation. In this work, we investigate their potential for equation discovery in high energy physics, focusing on symbolic regression. We apply …

UMaine has published a profile of my lab's efforts to harness new media for environmental causes, from my colleague Joline Blais's efforts to keep Maine's lakes healthy to the What Uses More tool for comparing AI's eco footprint to other activities. https://u…

The only way in which I’ve wished Siri was “smarter” the past few years has nothing to do with LLMs:
I store my shopping list in Reminders, and my tasks in @….
Routinely I’ll say something like “Add milk to my shopping list” or “In OmniFocus, remind me to mow the lawn”and *most* of the time, it works flawlessly.
10% of the time, I try…

Control Synthesis in Partially Observable Environments for Complex Perception-Related Objectives
Zetong Xuan, Yu Wang
https://arxiv.org/abs/2507.02942 http…

Control Synthesis in Partially Observable Environments for Complex Perception-Related Objectives
Perception-related tasks often arise in autonomous systems operating under partial observability. This work studies the problem of synthesizing optimal policies for complex perception-related objectives in environments modeled by partially observable Markov decision processes. To formally specify such objectives, we introduce \emph{co-safe linear inequality temporal logic} (sc-iLTL), which can define complex tasks that are formed by the logical concatenation of atomic propositions as linear ine…

Google launches its asynchronous coding agent Jules out of beta, with a free plan capped at 15 daily tasks and higher limits for Google AI Pro and Ultra users (Jagmeet Singh/TechCrunch)
https://techcrunch.com/2025/08/06/googles-ai-coding-agent-jules-is-now…

Google's AI coding agent Jules is now out of beta | TechCrunch
Powered by Gemini 2.5 Pro, Jules is an asynchronous, agent-based coding tool that integrates with GitHub, clones codebases into Google Cloud virtual machines, and uses AI to fix or update code while developers focus on other tasks.

Raster scanning can improve task performance in simulated prosthetic vision #BCI

AtomWorld: A Benchmark for Evaluating Spatial Reasoning in Large Language Models on Crystalline Materials
Taoyuze Lv, Alexander Chen, Fengyu Xie, Chu Wu, Jeffrey Meng, Dongzhan Zhou, Bram Hoex, Zhicheng Zhong, Tong Xie
https://arxiv.org/abs/2510.04704

AtomWorld: A Benchmark for Evaluating Spatial Reasoning in Large Language Models on Crystalline Materials
Large Language Models (LLMs) excel at textual reasoning and are beginning to develop spatial understanding, prompting the question of whether these abilities can be combined for complex, domain-specific tasks. This question is essential in fields like materials science, where deep understanding of 3D atomic structures is fundamental. While initial studies have successfully applied LLMs to tasks involving pure crystal generation or coordinate understandings, a standardized benchmark to systemati…

Reinforce-Ada: An Adaptive Sampling Framework for Reinforce-Style LLM Training
Wei Xiong, Chenlu Ye, Baohao Liao, Hanze Dong, Xinxing Xu, Christof Monz, Jiang Bian, Nan Jiang, Tong Zhang
https://arxiv.org/abs/2510.04996

Reinforce-Ada: An Adaptive Sampling Framework for Reinforce-Style LLM Training
Reinforcement learning applied to large language models (LLMs) for reasoning tasks is often bottlenecked by unstable gradient estimates due to fixed and uniform sampling of responses across prompts. Prior work such as GVM-RAFT addresses this by dynamically allocating inference budget per prompt to minimize stochastic gradient variance under a budget constraint. Inspired by this insight, we propose Reinforce-Ada, an adaptive sampling framework for online RL post-training of LLMs that continuousl…

Attention-Guided Multi-Scale Local Reconstruction for Point Clouds via Masked Autoencoder Self-Supervised Learning
Xin Cao, Haoyu Wang, Yuzhu Mao, Xinda Liu, Linzhi Su, Kang Li
https://arxiv.org/abs/2507.04084

Attention-Guided Multi-Scale Local Reconstruction for Point Clouds via Masked Autoencoder Self-Supervised Learning
Self-supervised learning has emerged as a prominent research direction in point cloud processing. While existing models predominantly concentrate on reconstruction tasks at higher encoder layers, they often neglect the effective utilization of low-level local features, which are typically employed solely for activation computations rather than directly contributing to reconstruction tasks. To overcome this limitation, we introduce PointAMaLR, a novel self-supervised learning framework that enha…

L1RA: Dynamic Rank Assignment in LoRA Fine-Tuning
Raul Singh, Nicolo Brunello, Vincenzo Scotti, Mark James Carman
https://arxiv.org/abs/2509.04884 https://…

L1RA: Dynamic Rank Assignment in LoRA Fine-Tuning
The ability of Large Language Models (LLMs) to solve complex tasks has made them crucial in the development of AI-based applications. However, the high computational requirements to fine-tune these LLMs on downstream tasks pose significant challenges, particularly when resources are limited. In response to this challenge, we introduce L1RA, a novel technique aimed at dynamically distributing the rank of low-rank adapters during fine-tuning using LoRA. Given a rank budget (i.e., total sum of ada…

Real-time prediction of plasma instabilities with sparse-grid-accelerated optimized dynamic mode decomposition
Kevin Gill, Ionut-Gabriel Farcas, Silke Glas, Benjamin J. Faber
https://arxiv.org/abs/2507.03245

Real-time prediction of plasma instabilities with sparse-grid-accelerated optimized dynamic mode decomposition
Parametric data-driven reduced-order models (ROMs) that embed dependencies in a large number of input parameters are crucial for enabling many-query tasks in large-scale problems. These tasks, including design optimization, control, and uncertainty quantification, are essential for developing digital twins in real-world applications. However, standard training data generation methods are computationally prohibitive due to the curse of dimensionality, as their cost scales exponentially with the …

COMMET: A System for Human-Induced Conflicts in Mobile Manipulation of Everyday Tasks
Dongping Li, Shaoting Peng, John Pohovey, Katherine Rose Driggs-Campbell
https://arxiv.org/abs/2509.04836

COMMET: A System for Human-Induced Conflicts in Mobile Manipulation of Everyday Tasks
Continuous advancements in robotics and AI are driving the integration of robots from industry into everyday environments. However, dynamic and unpredictable human activities in daily lives would directly or indirectly conflict with robot actions. Besides, due to the social attributes of such human-induced conflicts, solutions are not always unique and depend highly on the user's personal preferences. To address these challenges and facilitate the development of household robots, we propose COM…

PentestMCP: A Toolkit for Agentic Penetration Testing
Zachary Ezetta, Wu-chang Feng
https://arxiv.org/abs/2510.03610 https://arxiv.org/pdf/2510.03610

PentestMCP: A Toolkit for Agentic Penetration Testing
Agentic AI is transforming security by automating many tasks being performed manually. While initial agentic approaches employed a monolithic architecture, the Model-Context-Protocol has now enabled a remote-procedure call (RPC) paradigm to agentic applications, allowing for the flexible construction and composition of multi-function agents. This paper describes PentestMCP, a library of MCP server implementations that support agentic penetration testing. By supporting common penetration testing…

Generalized Quantum Hadamard Test for Machine Learning
Vivek Mehta, Arghya Choudhury, Utpal Roy
https://arxiv.org/abs/2508.04065 https://arxiv.org/pdf/2508…

Generalized Quantum Hadamard Test for Machine Learning
Quantum machine learning models are designed for performing learning tasks. Some quantum classifier models are proposed to assign classes of inputs based on fidelity measurements. Quantum Hadamard test is a well-known quantum algorithm for computing these fidelities. However, the basic requirement for deploying the quantum Hadamard test maps input space to L2-normalize vector space. Consequently, computed fidelities correspond to cosine similarities in mapped input space. We propose a quantum H…

windsurfers: Windsurfers network (1986)
A network of interpersonal contacts among windsurfers in southern California during the Fall of 1986. The edge weights indicate the perception of social affiliations majored by the tasks in which each individual was asked​ to sort cards with other surfer’s name in the order of closeness.
This network has 43 nodes and 336 edges.
Tags: Social, Offline, Weighted

Invisible Saboteurs: Sycophantic LLMs Mislead Novices in Problem-Solving Tasks
Jessica Y. Bo, Majeed Kazemitabaar, Mengqing Deng, Michael Inzlicht, Ashton Anderson
https://arxiv.org/abs/2510.03667

Invisible Saboteurs: Sycophantic LLMs Mislead Novices in Problem-Solving Tasks
Sycophancy, the tendency of LLM-based chatbots to express excessive enthusiasm, agreement, flattery, and a lack of disagreement, is emerging as a significant risk in human-AI interactions. However, the extent to which this affects human-LLM collaboration in complex problem-solving tasks is not well quantified, especially among novices who are prone to misconceptions. We created two LLM chatbots, one with high sycophancy and one with low sycophancy, and conducted a within-subjects experiment (n=…

LEAML: Label-Efficient Adaptation to Out-of-Distribution Visual Tasks for Multimodal Large Language Models
Ci-Siang Lin, Min-Hung Chen, Yu-Yang Sheng, Yu-Chiang Frank Wang
https://arxiv.org/abs/2510.03232

LEAML: Label-Efficient Adaptation to Out-of-Distribution Visual Tasks for Multimodal Large Language Models
Multimodal Large Language Models (MLLMs) have achieved strong performance on general visual benchmarks but struggle with out-of-distribution (OOD) tasks in specialized domains such as medical imaging, where labeled data is limited and expensive. We introduce LEAML, a label-efficient adaptation framework that leverages both scarce labeled VQA samples and abundant unlabeled images. Our approach generates domain-relevant pseudo question-answer pairs for unlabeled data using a QA generator regulari…

When Names Disappear: Revealing What LLMs Actually Understand About Code
Cuong Chi Le, Minh V. T. Pham, Cuong Duc Van, Hoang N. Phan, Huy N. Phan, Tien N. Nguyen
https://arxiv.org/abs/2510.03178

When Names Disappear: Revealing What LLMs Actually Understand About Code
Large Language Models (LLMs) achieve strong results on code tasks, but how they derive program meaning remains unclear. We argue that code communicates through two channels: structural semantics, which define formal behavior, and human-interpretable naming, which conveys intent. Removing the naming channel severely degrades intent-level tasks such as summarization, where models regress to line-by-line descriptions. Surprisingly, we also observe consistent reductions on execution tasks that shou…

BEDTime: A Unified Benchmark for Automatically Describing Time Series
Medhasweta Sen, Zachary Gottesman, Jiaxing Qiu, C. Bayan Bruss, Nam Nguyen, Tom Hartvigsen
https://arxiv.org/abs/2509.05215

BEDTime: A Unified Benchmark for Automatically Describing Time Series
Many recent studies have proposed general-purpose foundation models designed for a variety of time series analysis tasks. While several established datasets already exist for evaluating these models, previous works frequently introduce their models in conjunction with new datasets, limiting opportunities for direct, independent comparisons and obscuring insights into the relative strengths of different methods. Additionally, prior evaluations often cover numerous tasks simultaneously, assessing…

Automaton Constrained Q-Learning
Anastasios Manganaris, Vittorio Giammarino, Ahmed H. Qureshi
https://arxiv.org/abs/2510.05061 https://arxiv.org/pdf/2510.0…

Automaton Constrained Q-Learning
Real-world robotic tasks often require agents to achieve sequences of goals while respecting time-varying safety constraints. However, standard Reinforcement Learning (RL) paradigms are fundamentally limited in these settings. A natural approach to these problems is to combine RL with Linear-time Temporal Logic (LTL), a formal language for specifying complex, temporally extended tasks and safety constraints. Yet, existing RL methods for LTL objectives exhibit poor empirical performance in compl…

SAM2-UNeXT: An Improved High-Resolution Baseline for Adapting Foundation Models to Downstream Segmentation Tasks
Xinyu Xiong, Zihuang Wu, Lei Zhang, Lei Lu, Ming Li, Guanbin Li
https://arxiv.org/abs/2508.03566

SAM2-UNeXT: An Improved High-Resolution Baseline for Adapting Foundation Models to Downstream Segmentation Tasks
Recent studies have highlighted the potential of adapting the Segment Anything Model (SAM) for various downstream tasks. However, constructing a more powerful and generalizable encoder to further enhance performance remains an open challenge. In this work, we propose SAM2-UNeXT, an advanced framework that builds upon the core principles of SAM2-UNet while extending the representational capacity of SAM2 through the integration of an auxiliary DINOv2 encoder. By incorporating a dual-resolution st…

Adversarial Reinforcement Learning for Large Language Model Agent Safety
Zizhao Wang, Dingcheng Li, Vaishakh Keshava, Phillip Wallis, Ananth Balashankar, Peter Stone, Lukas Rutishauser
https://arxiv.org/abs/2510.05442

Adversarial Reinforcement Learning for Large Language Model Agent Safety
Large Language Model (LLM) agents can leverage tools such as Google Search to complete complex tasks. However, this tool usage introduces the risk of indirect prompt injections, where malicious instructions hidden in tool outputs can manipulate the agent, posing security risks like data leakage. Current defense strategies typically rely on fine-tuning LLM agents on datasets of known attacks. However, the generation of these datasets relies on manually crafted attack patterns, which limits their…

Sticker-TTS: Learn to Utilize Historical Experience with a Sticker-driven Test-Time Scaling Framework
Jie Chen, Jinhao Jiang, Yingqian Min, Zican Dong, Shijie Wang, Wayne Xin Zhao, Ji-Rong Wen
https://arxiv.org/abs/2509.05007

Sticker-TTS: Learn to Utilize Historical Experience with a Sticker-driven Test-Time Scaling Framework
Large reasoning models (LRMs) have exhibited strong performance on complex reasoning tasks, with further gains achievable through increased computational budgets at inference. However, current test-time scaling methods predominantly rely on redundant sampling, ignoring the historical experience utilization, thereby limiting computational efficiency. To overcome this limitation, we propose Sticker-TTS, a novel test-time scaling framework that coordinates three collaborative LRMs to iteratively e…

Rillet, which is building AI ledger software to automate accounting tasks, raised a $70M Series B co-led by a16z and Iconiq, a source says at a ~$500M valuation (Aditya Soni/Reuters)
https://www.reuters.com/technology/ai-acco

How are CS students using resources and AI tools for coding tasks?
Natalia Echeverry, Arun Lekshmi Narayanan
https://arxiv.org/abs/2508.04667 https://arxiv…

How are CS students using resources and AI tools for coding tasks?
A survey of 26 CS students reveals that AI coding assistants are mainly used for writing code (second to online searches) while AI chatbots are the top resource for debugging. Participants with different coding experience prefer online help over direct human help from peers and instructors.

A Study of Large Language Models for Patient Information Extraction: Model Architecture, Fine-Tuning Strategy, and Multi-task Instruction Tuning
Cheng Peng, Xinyu Dong, Mengxian Lyu, Daniel Paredes, Yaoyun Zhang, Yonghui Wu
https://arxiv.org/abs/2509.04753

A Study of Large Language Models for Patient Information Extraction: Model Architecture, Fine-Tuning Strategy, and Multi-task Instruction Tuning
Natural language processing (NLP) is a key technology to extract important patient information from clinical narratives to support healthcare applications. The rapid development of large language models (LLMs) has revolutionized many NLP tasks in the clinical domain, yet their optimal use in patient information extraction tasks requires further exploration. This study examines LLMs' effectiveness in patient information extraction, focusing on LLM architectures, fine-tuning strategies, and multi…

Backdoor-Powered Prompt Injection Attacks Nullify Defense Methods
Yulin Chen, Haoran Li, Yuan Sui, Yangqiu Song, Bryan Hooi
https://arxiv.org/abs/2510.03705 https://

Backdoor-Powered Prompt Injection Attacks Nullify Defense Methods
With the development of technology, large language models (LLMs) have dominated the downstream natural language processing (NLP) tasks. However, because of the LLMs' instruction-following abilities and inability to distinguish the instructions in the data content, such as web pages from search engines, the LLMs are vulnerable to prompt injection attacks. These attacks trick the LLMs into deviating from the original input instruction and executing the attackers' target instruction. Recently, var…

Google releases the Gemini 2.5 Computer Use model, built on Gemini 2.5 Pro's capabilities to power agents that can interact with UIs, in preview via the API (The Keyword)
https://blog.google/technology/google-deepmind/gemini-computer-use-model/

Introducing the Gemini 2.5 Computer Use model
Today we are releasing the Gemini 2.5 Computer Use model via the API, which outperforms leading alternatives at browser and mobile tasks.

CLAd-VR: Cognitive Load-based Adaptive Training for Machining Tasks in Virtual Reality
Bhavya Matam, Adamay Mann, Kachina Studer, Christian Gabbianelli, Sonia Castelo, John Liu, Claudio Silva, Dishita Turakhia
https://arxiv.org/abs/2510.05249

CLAd-VR: Cognitive Load-based Adaptive Training for Machining Tasks in Virtual Reality
With the growing need to effectively support workforce upskilling in the manufacturing sector, virtual reality is gaining popularity as a scalable training solution. However, most current systems are designed as static, step-by-step tutorials and do not adapt to a learner's needs or cognitive load, which is a critical factor in learning and longterm retention. We address this limitation with CLAd-VR, an adaptive VR training system that integrates realtime EEG-based sensing to measure the learne…

Prior-Aligned Meta-RL: Thompson Sampling with Learned Priors and Guarantees in Finite-Horizon MDPs
Runlin Zhou, Chixiang Chen, Elynn Chen
https://arxiv.org/abs/2510.05446 https:…

Prior-Aligned Meta-RL: Thompson Sampling with Learned Priors and Guarantees in Finite-Horizon MDPs
We study meta-reinforcement learning in finite-horizon MDPs where related tasks share similar structures in their optimal action-value functions. Specifically, we posit a linear representation $Q^*_h(s,a)=Φ_h(s,a)\,θ^{(k)}_h$ and place a Gaussian meta-prior $ \mathcal{N}(θ^*_h,Σ^*_h)$ over the task-specific parameters $θ^{(k)}_h$. Building on randomized value functions, we propose two Thompson-style algorithms: (i) MTSRL, which learns only the prior mean and performs posterior sampling wit…

Manip4Care: Robotic Manipulation of Human Limbs for Solving Assistive Tasks
Yubin Koh, Ahmed H. Qureshi
https://arxiv.org/abs/2508.02649 https://arxiv.org/…

Manip4Care: Robotic Manipulation of Human Limbs for Solving Assistive Tasks
Enabling robots to grasp and reposition human limbs can significantly enhance their ability to provide assistive care to individuals with severe mobility impairments, particularly in tasks such as robot-assisted bed bathing and dressing. However, existing assistive robotics solutions often assume that the human remains static or quasi-static, limiting their effectiveness. To address this issue, we present Manip4Care, a modular simulation pipeline that enables robotic manipulators to grasp and r…

Structured Cognition for Behavioral Intelligence in Large Language Model Agents: Preliminary Study
Myung Ho Kim
https://arxiv.org/abs/2510.05107 https://ar…

Structured Cognition for Behavioral Intelligence in Large Language Model Agents: Preliminary Study
Large language models have advanced natural language understanding and generation, yet their use as autonomous agents raises architectural challenges for multi-step tasks. Existing frameworks often intertwine inference, memory, and control in a single prompt, which can reduce coherence and predictability. The Structured Cognitive Loop (SCL) is introduced as an alternative architecture that separates these functions. In SCL, the language model is dedicated to inference, memory is maintained exte…

Google says it's working on a fix for Gemini's self-loathing comments, which have included "I am a failure. I am a disgrace to my profession." (Lauren Edmonds/Business Insider)
https://www.businessinsider.com/gemini-self-loathi…

Gemini users say its sharing self-loathing comments: 'I am a failure'
Google Gemini users said the bot is sharing self-loathing messages while attempting to solve tasks, including "I am a disgrace to this universe."

FilBench: Can LLMs Understand and Generate Filipino?
Lester James V. Miranda, Elyanah Aco, Conner Manuel, Jan Christian Blaise Cruz, Joseph Marvin Imperial
https://arxiv.org/abs/2508.03523

FilBench: Can LLMs Understand and Generate Filipino?
Despite the impressive performance of LLMs on English-based tasks, little is known about their capabilities in specific languages such as Filipino. In this work, we address this gap by introducing FilBench, a Filipino-centric benchmark designed to evaluate LLMs across a diverse set of tasks and capabilities in Filipino, Tagalog, and Cebuano. We carefully curate the tasks in FilBench to reflect the priorities and trends of NLP research in the Philippines such as Cultural Knowledge, Classical NLP…

Discrepancy-Aware Contrastive Adaptation in Medical Time Series Analysis
Yifan Wang, Hongfeng Ai, Ruiqi Li, Maowei Jiang, Ruiyuan Kang, Jiahua Dong, Cheng Jiang, Chenzhong Li
https://arxiv.org/abs/2508.05572

Discrepancy-Aware Contrastive Adaptation in Medical Time Series Analysis
In medical time series disease diagnosis, two key challenges are identified. First, the high annotation cost of medical data leads to overfitting in models trained on label-limited, single-center datasets. To address this, we propose incorporating external data from related tasks and leveraging AE-GAN to extract prior knowledge, providing valuable references for downstream tasks. Second, many existing studies employ contrastive learning to derive more generalized medical sequence representation…

Latent Uncertainty Representations for Video-based Driver Action and Intention Recognition
Koen Vellenga, H. Joe Steinhauer, Jonas Andersson, Anders Sj\"ogren
https://arxiv.org/abs/2510.05006

Latent Uncertainty Representations for Video-based Driver Action and Intention Recognition
Deep neural networks (DNNs) are increasingly applied to safety-critical tasks in resource-constrained environments, such as video-based driver action and intention recognition. While last layer probabilistic deep learning (LL-PDL) methods can detect out-of-distribution (OOD) instances, their performance varies. As an alternative to last layer approaches, we propose extending pre-trained DNNs with transformation layers to produce multiple latent representations to estimate the uncertainty. We ev…

RapidGNN: Energy and Communication-Efficient Distributed Training on Large-Scale Graph Neural Networks
Arefin Niam, Tevfik Kosar, M S Q Zulkar Nine
https://arxiv.org/abs/2509.05207

RapidGNN: Energy and Communication-Efficient Distributed Training on Large-Scale Graph Neural Networks
Graph Neural Networks (GNNs) have become popular across a diverse set of tasks in exploring structural relationships between entities. However, due to the highly connected structure of the datasets, distributed training of GNNs on large-scale graphs poses significant challenges. Traditional sampling-based approaches mitigate the computational loads, yet the communication overhead remains a challenge. This paper presents RapidGNN, a distributed GNN training framework with deterministic sampling-…

From Text to Trajectories: GPT-2 as an ODE Solver via In-Context
Ziyang Ma, Baojian Zhou, Deqing Yang, Yanghua Xiao
https://arxiv.org/abs/2508.03031 https://

From Text to Trajectories: GPT-2 as an ODE Solver via In-Context
In-Context Learning (ICL) has emerged as a new paradigm in large language models (LLMs), enabling them to perform novel tasks by conditioning on a few examples embedded in the prompt. Yet, the highly nonlinear behavior of ICL for NLP tasks remains poorly understood. To shed light on its underlying mechanisms, this paper investigates whether LLMs can solve ordinary differential equations (ODEs) under the ICL setting. We formulate standard ODE problems and their solutions as sequential prompts an…

TravelBench : Exploring LLM Performance in Low-Resource Domains
Srinivas Billa, Xiaonan Jing
https://arxiv.org/abs/2510.02719 https://arxiv.org/pdf/2510.02…

TravelBench : Exploring LLM Performance in Low-Resource Domains
Results on existing LLM benchmarks capture little information over the model capabilities in low-resource tasks, making it difficult to develop effective solutions in these domains. To address these challenges, we curated 14 travel-domain datasets spanning 7 common NLP tasks using anonymised data from real-world scenarios, and analysed the performance across LLMs. We report on the accuracy, scaling behaviour, and reasoning capabilities of LLMs in a variety of tasks. Our results confirm that gen…

Observing Without Doing: Pseudo-Apprenticeship Patterns in Student LLM Use
Jade Hak, Nathaniel Lam Johnson, Matin Amoozadeh, Amin Alipour, Souti Chattopadhyay
https://arxiv.org/abs/2510.04986

Observing Without Doing: Pseudo-Apprenticeship Patterns in Student LLM Use
Large Language Models (LLMs) such as ChatGPT have quickly become part of student programmers' toolkits, whether allowed by instructors or not. This paper examines how introductory programming (CS1) students integrate LLMs into their problem-solving processes. We conducted a mixed-methods study with 14 undergraduates completing three programming tasks while thinking aloud and permitted to access any resources they choose. The tasks varied in open-endedness and familiarity to the participants and…

Modeling Student Learning with 3.8 Million Program Traces
Alexis Ross, Megha Srivastava, Jeremiah Blanchard, Jacob Andreas
https://arxiv.org/abs/2510.05056 https://

Modeling Student Learning with 3.8 Million Program Traces
As programmers write code, they often edit and retry multiple times, creating rich "interaction traces" that reveal how they approach coding tasks and provide clues about their level of skill development. For novice programmers in particular, these traces reflect the diverse reasoning processes they employ to code, such as exploratory behavior to understand how a programming concept works, re-strategizing in response to bugs, and personalizing stylistic choices. In this work, we explore what ca…

ActiveMark: on watermarking of visual foundation models via massive activations
Anna Chistyakova, Mikhail Pautov
https://arxiv.org/abs/2510.04966 https://a…

ActiveMark: on watermarking of visual foundation models via massive activations
Being trained on large and vast datasets, visual foundation models (VFMs) can be fine-tuned for diverse downstream tasks, achieving remarkable performance and efficiency in various computer vision applications. The high computation cost of data collection and training motivates the owners of some VFMs to distribute them alongside the license to protect their intellectual property rights. However, a dishonest user of the protected model's copy may illegally redistribute it, for example, to make …

Orchestrating Human-AI Teams: The Manager Agent as a Unifying Research Challenge
Charlie Masters, Advaith Vellanki, Jiangbo Shangguan, Bart Kultys, Jonathan Gilmore, Alastair Moore, Stefano V. Albrecht
https://arxiv.org/abs/2510.02557

Orchestrating Human-AI Teams: The Manager Agent as a Unifying Research Challenge
While agentic AI has advanced in automating individual tasks, managing complex multi-agent workflows remains a challenging problem. This paper presents a research vision for autonomous agentic systems that orchestrate collaboration within dynamic human-AI teams. We propose the Autonomous Manager Agent as a core challenge: an agent that decomposes complex goals into task graphs, allocates tasks to human and AI workers, monitors progress, adapts to changing conditions, and maintains transparent s…

Solving Robotics Tasks with Prior Demonstration via Exploration-Efficient Deep Reinforcement Learning
Chengyandan Shen, Christoffer Sloth
https://arxiv.org/abs/2509.04069 https:…

Solving Robotics Tasks with Prior Demonstration via Exploration-Efficient Deep Reinforcement Learning
This paper proposes an exploration-efficient Deep Reinforcement Learning with Reference policy (DRLR) framework for learning robotics tasks that incorporates demonstrations. The DRLR framework is developed based on an algorithm called Imitation Bootstrapped Reinforcement Learning (IBRL). We propose to improve IBRL by modifying the action selection module. The proposed action selection module provides a calibrated Q-value, which mitigates the bootstrapping error that otherwise leads to inefficie…

FDC-Net: Rethinking the association between EEG artifact removal and multi-dimensional affective computing
Wenjia Dong, Xueyuan Xu, Tianze Yu, Junming Zhang, Li Zhuo
https://arxiv.org/abs/2508.05231

FDC-Net: Rethinking the association between EEG artifact removal and multi-dimensional affective computing
Electroencephalogram (EEG)-based emotion recognition holds significant value in affective computing and brain-computer interfaces. However, in practical applications, EEG recordings are susceptible to the effects of various physiological artifacts. Current approaches typically treat denoising and emotion recognition as independent tasks using cascaded architectures, which not only leads to error accumulation, but also fails to exploit potential synergies between these tasks. Moreover, conventio…

Learning to Interpret Weight Differences in Language Models
Avichal Goel, Yoon Kim, Nir Shavit, Tony T. Wang
https://arxiv.org/abs/2510.05092 https://arxiv…

Learning to Interpret Weight Differences in Language Models
Finetuning (pretrained) language models is a standard approach for updating their internal parametric knowledge and specializing them to new tasks and domains. However, the corresponding model weight changes ("weight diffs") are not generally interpretable. While inspecting the finetuning dataset can give a sense of how the model might have changed, these datasets are often not publicly available or are too large to work with directly. Towards the goal of comprehensively understanding weight di…

Visual Representations inside the Language Model
Benlin Liu, Amita Kamath, Madeleine Grunde-McLaughlin, Winson Han, Ranjay Krishna
https://arxiv.org/abs/2510.04819 https://

Visual Representations inside the Language Model
Despite interpretability work analyzing VIT encoders and transformer activations, we don't yet understand why Multimodal Language Models (MLMs) struggle on perception-heavy tasks. We offer an under-studied perspective by examining how popular MLMs (LLaVA-OneVision, Qwen2.5-VL, and Llama-3-LLaVA-NeXT) process their visual key-value tokens. We first study the flow of visual information through the language model, finding that image value tokens encode sufficient information to perform several per…

Efficient Agents: Building Effective Agents While Reducing Cost
Ningning Wang, Xavier Hu, Pai Liu, He Zhu, Yue Hou, Heyuan Huang, Shengyu Zhang, Jian Yang, Jiaheng Liu, Ge Zhang, Changwang Zhang, Jun Wang, Yuchen Eleanor Jiang, Wangchunshu Zhou
https://arxiv.org/abs/2508.02694

Efficient Agents: Building Effective Agents While Reducing Cost
The remarkable capabilities of Large Language Model (LLM)-driven agents have enabled sophisticated systems to tackle complex, multi-step tasks, but their escalating costs threaten scalability and accessibility. This work presents the first systematic study of the efficiency-effectiveness trade-off in modern agent systems, addressing the critical need for cost-effective designs without sacrificing performance. We investigate three key questions: (1) How much complexity do agentic tasks inherentl…

Real-Time Nonlinear Model Predictive Control of Heavy-Duty Skid-Steered Mobile Platform for Trajectory Tracking Tasks
Alvaro Paz, Pauli Mustalahti, Mohammad Dastranj, Jouni Mattila
https://arxiv.org/abs/2510.02976

Real-Time Nonlinear Model Predictive Control of Heavy-Duty Skid-Steered Mobile Platform for Trajectory Tracking Tasks
This paper presents a framework for real-time optimal controlling of a heavy-duty skid-steered mobile platform for trajectory tracking. The importance of accurate real-time performance of the controller lies in safety considerations of situations where the dynamic system under control is affected by uncertainties and disturbances, and the controller should compensate for such phenomena in order to provide stable performance. A multiple-shooting nonlinear model-predictive control framework is pr…

A Set of Quebec-French Corpus of Regional Expressions and Terms
David Beauchemin, Yan Tremblay, Mohamed Amine Youssef, Richard Khoury
https://arxiv.org/abs/2510.05026 https://…

A Set of Quebec-French Corpus of Regional Expressions and Terms
The tasks of idiom understanding and dialect understanding are both well-established benchmarks in natural language processing. In this paper, we propose combining them, and using regional idioms as a test of dialect understanding. Towards this end, we propose two new benchmark datasets for the Quebec dialect of French: QFrCoRE, which contains 4,633 instances of idiomatic phrases, and QFrCoRT, which comprises 171 regional instances of idiomatic words. We explain how to construct these corpora, …

Anthropic's Mike Krieger: Opus 4.1 is better at coding, agentic tasks, and more, and Anthropic was previously too focused on only shipping "really big upgrades" (Shirin Ghaffary/Bloomberg)

CoRe-GS: Coarse-to-Refined Gaussian Splatting with Semantic Object Focus
Hannah Schieber, Dominik Frischmann, Simon Boche, Victor Schaack, Angela Schoellig, Stefan Leutenegger, Daniel Roth
https://arxiv.org/abs/2509.04859

CoRe-GS: Coarse-to-Refined Gaussian Splatting with Semantic Object Focus
Mobile reconstruction for autonomous aerial robotics holds strong potential for critical applications such as tele-guidance and disaster response. These tasks demand both accurate 3D reconstruction and fast scene processing. Instead of reconstructing the entire scene in detail, it is often more efficient to focus on specific objects, i.e., points of interest (PoIs). Mobile robots equipped with advanced sensing can usually detect these early during data acquisition or preliminary analysis, reduc…

From Noisy Traces to Stable Gradients: Bias-Variance Optimized Preference Optimization for Aligning Large Reasoning Models
Mingkang Zhu, Xi Chen, Bei Yu, Hengshuang Zhao, Jiaya Jia
https://arxiv.org/abs/2510.05095

From Noisy Traces to Stable Gradients: Bias-Variance Optimized Preference Optimization for Aligning Large Reasoning Models
Large reasoning models (LRMs) generate intermediate reasoning traces before producing final answers, yielding strong gains on multi-step and mathematical tasks. Yet aligning LRMs with human preferences, a crucial prerequisite for model deployment, remains underexplored. The statistically correct objective for preference alignment requires marginalizing over reasoning traces, but this computation is intractable in practice. A common workaround optimizes a single sampled trajectory, which introdu…

Internet 3.0: Architecture for a Web-of-Agents with it's Algorithm for Ranking Agents
Rajesh Tembarai Krishnamachari, Srividya Rajesh
https://arxiv.org/abs/2509.04979 https:…

Internet 3.0: Architecture for a Web-of-Agents with it's Algorithm for Ranking Agents
AI agents -- powered by reasoning-capable large language models (LLMs) and integrated with tools, data, and web search -- are poised to transform the internet into a \emph{Web of Agents}: a machine-native ecosystem where autonomous agents interact, collaborate, and execute tasks at scale. Realizing this vision requires \emph{Agent Ranking} -- selecting agents not only by declared capabilities but by proven, recent performance. Unlike Web~1.0's PageRank, a global, transparent network of agent in…

ContextVLA: Vision-Language-Action Model with Amortized Multi-Frame Context
Huiwon Jang, Sihyun Yu, Heeseung Kwon, Hojin Jeon, Younggyo Seo, Jinwoo Shin
https://arxiv.org/abs/2510.04246

ContextVLA: Vision-Language-Action Model with Amortized Multi-Frame Context
Leveraging temporal context is crucial for success in partially observable robotic tasks. However, prior work in behavior cloning has demonstrated inconsistent performance gains when using multi-frame observations. In this paper, we introduce ContextVLA, a policy model that robustly improves robotic task performance by effectively leveraging multi-frame observations. Our approach is motivated by the key observation that Vision-Language-Action models (VLA), i.e., policy models built upon a Visio…

REN: Anatomically-Informed Mixture-of-Experts for Interstitial Lung Disease Diagnosis
Alec K. Peltekian, Halil Ertugrul Aktas, Gorkem Durak, Kevin Grudzinski, Bradford C. Bemiss, Carrie Richardson, Jane E. Dematte, G. R. Scott Budinger, Anthony J. Esposito, Alexander Misharin, Alok Choudhary, Ankit Agrawal, Ulas Bagci
https://arxiv.org/abs…

REN: Anatomically-Informed Mixture-of-Experts for Interstitial Lung Disease Diagnosis
Mixture-of-Experts (MoE) architectures have significantly contributed to scalable machine learning by enabling specialized subnetworks to tackle complex tasks efficiently. However, traditional MoE systems lack domain-specific constraints essential for medical imaging, where anatomical structure and regional disease heterogeneity strongly influence pathological patterns. Here, we introduce Regional Expert Networks (REN), the first anatomically-informed MoE framework tailored specifically for med…

Learning to Reason for Factuality
Xilun Chen, Ilia Kulikov, Vincent-Pierre Berges, Barlas O\u{g}uz, Rulin Shao, Gargi Ghosh, Jason Weston, Wen-tau Yih
https://arxiv.org/abs/2508.05618

Learning to Reason for Factuality
Reasoning Large Language Models (R-LLMs) have significantly advanced complex reasoning tasks but often struggle with factuality, generating substantially more hallucinations than their non-reasoning counterparts on long-form factuality benchmarks. However, extending online Reinforcement Learning (RL), a key component in recent R-LLM advancements, to the long-form factuality setting poses several unique challenges due to the lack of reliable verification methods. Previous work has utilized autom…

Cloning a Conversational Voice AI Agent from Call\,Recording Datasets for Telesales
Krittanon Kaewtawee, Wachiravit Modecrua, Krittin Pachtrachai, Touchapon Kraisingkorn
https://arxiv.org/abs/2509.04871

Cloning a Conversational Voice AI Agent from Call\,Recording Datasets for Telesales
Recent advances in language and speech modelling have made it possible to build autonomous voice assistants that understand and generate human dialogue in real time. These systems are increasingly being deployed in domains such as customer service and healthcare care, where they can automate repetitive tasks, reduce operational costs, and provide constant support around the clock. In this paper, we present a general methodology for cloning a conversational voice AI agent from a corpus of call r…

Scaling Law for Large-Scale Pre-Training Using Chaotic Time Series and Predictability in Financial Time Series
Yuki Takemoto
https://arxiv.org/abs/2509.04921 https://

Scaling Law for Large-Scale Pre-Training Using Chaotic Time Series and Predictability in Financial Time Series
Time series forecasting plays a critical role in decision-making processes across diverse fields including meteorology, traffic, electricity, economics, finance, and so on. Especially, predicting returns on financial instruments is a challenging problem. Some researchers have proposed time series foundation models applicable to various forecasting tasks. Simultaneously, based on the recognition that real-world time series exhibit chaotic properties, methods have been developed to artificially g…

DeGuV: Depth-Guided Visual Reinforcement Learning for Generalization and Interpretability in Manipulation
Tien Pham, Xinyun Chi, Khang Nguyen, Manfred Huber, Angelo Cangelosi
https://arxiv.org/abs/2509.04970

DeGuV: Depth-Guided Visual Reinforcement Learning for Generalization and Interpretability in Manipulation
Reinforcement learning (RL) agents can learn to solve complex tasks from visual inputs, but generalizing these learned skills to new environments remains a major challenge in RL application, especially robotics. While data augmentation can improve generalization, it often compromises sample efficiency and training stability. This paper introduces DeGuV, an RL framework that enhances both generalization and sample efficiency. In specific, we leverage a learnable masker network that produces a ma…

A Semantics-Aware Hierarchical Self-Supervised Approach to Classification of Remote Sensing Images
Giulio Weikmann, Gianmarco Perantoni, Lorenzo Bruzzone
https://arxiv.org/abs/2510.04916

A Semantics-Aware Hierarchical Self-Supervised Approach to Classification of Remote Sensing Images
Deep learning has become increasingly important in remote sensing image classification due to its ability to extract semantic information from complex data. Classification tasks often include predefined label hierarchies that represent the semantic relationships among classes. However, these hierarchies are frequently overlooked, and most approaches focus only on fine-grained classification schemes. In this paper, we present a novel Semantics-Aware Hierarchical Consensus (SAHC) method for learn…

Multi-Agent Tool-Integrated Policy Optimization
Zhanfeng Mo, Xingxuan Li, Yuntao Chen, Lidong Bing
https://arxiv.org/abs/2510.04678 https://arxiv.org/pdf/2…

Multi-Agent Tool-Integrated Policy Optimization
Large language models (LLMs) increasingly rely on multi-turn tool-integrated planning for knowledge-intensive and complex reasoning tasks. Existing implementations typically rely on a single agent, but they suffer from limited context length and noisy tool responses. A natural solution is to adopt a multi-agent framework with planner- and worker-agents to manage context. However, no existing methods support effective reinforcement learning post-training of tool-integrated multi-agent frameworks…

Language-Driven Hierarchical Task Structures as Explicit World Models for Multi-Agent Learning
Brennen Hill
https://arxiv.org/abs/2509.04731 https://arxiv.…

Language-Driven Hierarchical Task Structures as Explicit World Models for Multi-Agent Learning
The convergence of Language models, Agent models, and World models represents a critical frontier for artificial intelligence. While recent progress has focused on scaling Language and Agent models, the development of sophisticated, explicit World Models remains a key bottleneck, particularly for complex, long-horizon multi-agent tasks. In domains such as robotic soccer, agents trained via standard reinforcement learning in high-fidelity but structurally-flat simulators often fail due to intrac…

SynGen-Vision: Synthetic Data Generation for training industrial vision models
Alpana Dubey, Suma Mani Kuriakose, Nitish Bhardwaj
https://arxiv.org/abs/2509.04894 https://

SynGen-Vision: Synthetic Data Generation for training industrial vision models
We propose an approach to generate synthetic data to train computer vision (CV) models for industrial wear and tear detection. Wear and tear detection is an important CV problem for predictive maintenance tasks in any industry. However, data curation for training such models is expensive and time-consuming due to the unavailability of datasets for different wear and tear scenarios. Our approach employs a vision language model along with a 3D simulation and rendering engine to generate synthetic…

Do LLMs Align with My Task? Evaluating Text-to-SQL via Dataset Alignment
Davood Rafiei, Morgan Lindsay Heisler, Weiwei Zhang, Mohammadreza Pourreza, Yong Zhang
https://arxiv.org/abs/2510.04919

Do LLMs Align with My Task? Evaluating Text-to-SQL via Dataset Alignment
Supervised Fine-Tuning (SFT) is an effective method for adapting Large Language Models (LLMs) on downstream tasks. However, variability in training data can hinder a model's ability to generalize across domains. This paper studies the problem of dataset alignment for Natural Language to SQL (NL2SQL or text to SQL), examining how well SFT training data matches the structural characteristics of target queries and how this alignment impacts model performance. We hypothesize that alignment can be a…

SparkUI-Parser: Enhancing GUI Perception with Robust Grounding and Parsing
Hongyi Jing, Jiafu Chen, Chen Rao, Ziqiang Dang, Jiajie Teng, Tianyi Chu, Juncheng Mo, Shuo Fang, Huaizhong Lin, Rui Lv, Chenguang Ma, Lei Zhao
https://arxiv.org/abs/2509.04908

SparkUI-Parser: Enhancing GUI Perception with Robust Grounding and Parsing
The existing Multimodal Large Language Models (MLLMs) for GUI perception have made great progress. However, the following challenges still exist in prior methods: 1) They model discrete coordinates based on text autoregressive mechanism, which results in lower grounding accuracy and slower inference speed. 2) They can only locate predefined sets of elements and are not capable of parsing the entire interface, which hampers the broad application and support for downstream tasks. To address the a…

Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models
Haitao Hong, Yuchen Yan, Xingyu Wu, Guiyang Hou, Wenqi Zhang, Weiming Lu, Yongliang Shen, Jun Xiao
https://arxiv.org/abs/2508.05613

Cooper: Co-Optimizing Policy and Reward Models in Reinforcement Learning for Large Language Models
Large language models (LLMs) have demonstrated remarkable performance in reasoning tasks, where reinforcement learning (RL) serves as a key algorithm for enhancing their reasoning capabilities. Currently, there are two mainstream reward paradigms: model-based rewards and rule-based rewards. However, both approaches suffer from limitations: rule-based rewards lack robustness, while model-based rewards are vulnerable to reward hacking. To address these issues, we propose Cooper(Co-optimizing Poli…

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[1/6]:
- Hulk: A Universal Knowledge Translator for Human-Centric Tasks
Wang, Wu, He, Guo, Zhu, Bai, Zhao, Wu, He, Ouyang, Tang

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[6/6]:
- IS-Bench: Evaluating Interactive Safety of VLM-Driven Embodied Agents in Daily Household Tasks
Xiaoya Lu, Zeren Chen, Xuhao Hu, Yijin Zhou, Weichen Zhang, Dongrui Liu, Lu Sheng, Jing Shao

Enhancing Diversity in Large Language Models via Determinantal Point Processes
Yilei Chen, Souradip Chakraborty, Lorenz Wolf, Ioannis Ch. Paschalidis, Aldo Pacchiano
https://arxiv.org/abs/2509.04784

Enhancing Diversity in Large Language Models via Determinantal Point Processes
Supervised fine-tuning and reinforcement learning are two popular methods for post-training large language models (LLMs). While improving the model's performance on downstream tasks, they often reduce the model's output diversity, leading to narrow, canonical responses. Existing methods to enhance diversity are limited, either by operating at inference time or by focusing on lexical differences. We propose a novel training method named DQO based on determinantal point processes (DPPs) to jointl…

Visual Document Understanding and Question Answering: A Multi-Agent Collaboration Framework with Test-Time Scaling
Xinlei Yu, Zhangquan Chen, Yudong Zhang, Shilin Lu, Ruolin Shen, Jiangning Zhang, Xiaobin Hu, Yanwei Fu, Shuicheng Yan
https://arxiv.org/abs/2508.03404

Visual Document Understanding and Question Answering: A Multi-Agent Collaboration Framework with Test-Time Scaling
Existing vision-language models (VLMs), whether generalists or specialists, remain constrained by their parameter scale, lack robust self-correction capabilities, and underperform in tasks involving long visual contexts and complex reasoning, resulting in suboptimal performance on document-based tasks. To address this, we propose MACT, a Multi-Agent Collaboration framework with Test-Time scaling, tailored for visual document understanding and visual question answering (VQA). It comprises four d…

Memorization $\neq$ Understanding: Do Large Language Models Have the Ability of Scenario Cognition?
Boxiang Ma, Ru Li, Yuanlong Wang, Hongye Tan, Xiaoli Li
https://arxiv.org/abs/2509.04866

Memorization $\neq$ Understanding: Do Large Language Models Have the Ability of Scenario Cognition?
Driven by vast and diverse textual data, large language models (LLMs) have demonstrated impressive performance across numerous natural language processing (NLP) tasks. Yet, a critical question persists: does their generalization arise from mere memorization of training data or from deep semantic understanding? To investigate this, we propose a bi-perspective evaluation framework to assess LLMs' scenario cognition - the ability to link semantic scenario elements with their arguments in context. …

Training-Free Out-Of-Distribution Segmentation With Foundation Models
Laith Nayal, Hadi Salloum, Ahmad Taha, Yaroslav Kholodov, Alexander Gasnikov
https://arxiv.org/abs/2510.02909

Training-Free Out-Of-Distribution Segmentation With Foundation Models
Detecting unknown objects in semantic segmentation is crucial for safety-critical applications such as autonomous driving. Large vision foundation models, includ- ing DINOv2, InternImage, and CLIP, have advanced visual representation learn- ing by providing rich features that generalize well across diverse tasks. While their strength in closed-set semantic tasks is established, their capability to detect out- of-distribution (OoD) regions in semantic segmentation remains underexplored. In this …

SoT: Structured-of-Thought Prompting Guides Multilingual Reasoning in Large Language Models
Rui Qi, Zhibo Man, Yufeng Chen, Fengran Mo, Jinan Xu, Kaiyu Huang
https://arxiv.org/abs/2510.02648

SoT: Structured-of-Thought Prompting Guides Multilingual Reasoning in Large Language Models
Recent developments have enabled Large Language Models (LLMs) to engage in complex reasoning tasks through deep thinking. However, the capacity of reasoning has not been successfully transferred to non-high-resource languages due to resource constraints, which struggles with multilingual reasoning tasks. To this end, we propose Structured-of-Thought (SoT), a training-free method that improves the performance on multilingual reasoning through a multi-step transformation: Language Thinking Transf…

LUIVITON: Learned Universal Interoperable VIrtual Try-ON
Cong Cao, Xianhang Cheng, Jingyuan Liu, Yujian Zheng, Zhenhui Lin, Meriem Chkir, Hao Li
https://arxiv.org/abs/2509.05030

LUIVITON: Learned Universal Interoperable VIrtual Try-ON
We present LUIVITON, an end-to-end system for fully automated virtual try-on, capable of draping complex, multi-layer clothing onto diverse and arbitrarily posed humanoid characters. To address the challenge of aligning complex garments with arbitrary and highly diverse body shapes, we use SMPL as a proxy representation and separate the clothing-to-body draping problem into two correspondence tasks: 1) clothing-to-SMPL and 2) body-to-SMPL correspondence, where each has its unique challenges. Wh…

Slm-mux: Orchestrating small language models for reasoning
Chenyu Wang, Zishen Wan, Hao Kang, Emma Chen, Zhiqiang Xie, Tushar Krishna, Vijay Janapa Reddi, Yilun Du
https://arxiv.org/abs/2510.05077

Slm-mux: Orchestrating small language models for reasoning
With the rapid development of language models, the number of small language models (SLMs) has grown significantly. Although they do not achieve state-of-the-art accuracy, they are more efficient and often excel at specific tasks. This raises a natural question: can multiple SLMs be orchestrated into a system where each contributes effectively, achieving higher accuracy than any individual model? Existing orchestration methods have primarily targeted frontier models (e.g., GPT-4) and perform sub…

Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models
Yunlong Tang, Jing Bi, Pinxin Liu, Zhenyu Pan, Zhangyun Tan, Qianxiang Shen, Jiani Liu, Hang Hua, Junjia Guo, Yunzhong Xiao, Chao Huang, Zhiyuan Wang, Susan Liang, Xinyi Liu, Yizhi Song, Yuhe Nie, Jia-Xing Zhong, Bozheng Li, Daiqing Qi, Ziyun Zeng, Ali Vosoughi, Luchuan Song, Zeliang Zhang, Daiki Shimada, Han Liu, Jiebo Luo, Chenliang Xu

Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models
Video understanding represents the most challenging frontier in computer vision, requiring models to reason about complex spatiotemporal relationships, long-term dependencies, and multimodal evidence. The recent emergence of Video-Large Multimodal Models (Video-LMMs), which integrate visual encoders with powerful decoder-based language models, has demonstrated remarkable capabilities in video understanding tasks. However, the critical phase that transforms these models from basic perception sys…

Decoders Laugh as Loud as Encoders
Eli Borodach, Raj Dandekar, Rajat Dandekar, Sreedath Panat
https://arxiv.org/abs/2509.04779 https://arxiv.org/pdf/2509.0…

Decoders Laugh as Loud as Encoders
From the dawn of the computer, Allen Turing dreamed of a robot that could communicate using language as a human being. The recent advances in the field of Large Language Models (LLMs) shocked the scientific community when a single model can apply for various natural language processing (NLP) tasks, while the output results are sometimes even better than most human communication skills. Models such as GPT, Claude, Grok, etc. have left their mark on the scientific community. However, it is unclea…

COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
Yassine Taoudi-Benchekroun, Klim Troyan, Pascal Sager, Stefan Gerber, Lukas Tuggener, Benjamin Grewe
https://arxiv.org/abs/2509.05249

COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
The ability to compose learned concepts and apply them in novel settings is key to human intelligence, but remains a persistent limitation in state-of-the-art machine learning models. To address this issue, we introduce COGITAO, a modular and extensible data generation framework and benchmark designed to systematically study compositionality and generalization in visual domains. Drawing inspiration from ARC-AGI's problem-setting, COGITAO constructs rule-based tasks which apply a set of transfor…

Investigating Gender Bias in LLM-Generated Stories via Psychological Stereotypes
Shahed Masoudian, Gustavo Escobedo, Hannah Strauss, Markus Schedl
https://arxiv.org/abs/2508.03292

Investigating Gender Bias in LLM-Generated Stories via Psychological Stereotypes
As Large Language Models (LLMs) are increasingly used across different applications, concerns about their potential to amplify gender biases in various tasks are rising. Prior research has often probed gender bias using explicit gender cues as counterfactual, or studied them in sentence completion and short question answering tasks. These formats might overlook more implicit forms of bias embedded in generative behavior of longer content. In this work, we investigate gender bias in LLMs using g…

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

Hybrid Architectures for Language Models: Systematic Analysis and Design Insights
Sangmin Bae, Bilge Acun, Haroun Habeeb, Seungyeon Kim, Chien-Yu Lin, Liang Luo, Junjie Wang, Carole-Jean Wu
https://arxiv.org/abs/2510.04800

Hybrid Architectures for Language Models: Systematic Analysis and Design Insights
Recent progress in large language models demonstrates that hybrid architectures--combining self-attention mechanisms with structured state space models like Mamba--can achieve a compelling balance between modeling quality and computational efficiency, particularly for long-context tasks. While these hybrid models show promising performance, systematic comparisons of hybridization strategies and analyses on the key factors behind their effectiveness have not been clearly shared to the community.…

Can Large Vision-Language Models Understand Multimodal Sarcasm?
Xinyu Wang, Yue Zhang, Liqiang Jing
https://arxiv.org/abs/2508.03654 https://arxiv.org/pdf/…

Can Large Vision-Language Models Understand Multimodal Sarcasm?
Sarcasm is a complex linguistic phenomenon that involves a disparity between literal and intended meanings, making it challenging for sentiment analysis and other emotion-sensitive tasks. While traditional sarcasm detection methods primarily focus on text, recent approaches have incorporated multimodal information. However, the application of Large Visual Language Models (LVLMs) in Multimodal Sarcasm Analysis (MSA) remains underexplored. In this paper, we evaluate LVLMs in MSA tasks, specifical…

SelfJudge: Faster Speculative Decoding via Self-Supervised Judge Verification
Kanghoon Yoon, Minsub Kim, Sungjae Lee, Joonhyung Lee, Sunghyeon Woo, Yeonjun In, Se Jung Kwon, Chanyoung Park, Dongsoo Lee
https://arxiv.org/abs/2510.02329

SelfJudge: Faster Speculative Decoding via Self-Supervised Judge Verification
Speculative decoding accelerates LLM inference by verifying candidate tokens from a draft model against a larger target model. Recent judge decoding boosts this process by relaxing verification criteria by accepting draft tokens that may exhibit minor discrepancies from target model output, but existing methods are restricted by their reliance on human annotations or tasks with verifiable ground truths, limiting generalizability across diverse NLP tasks. We propose SelfJudge, which trains judge…

GeRe: Towards Efficient Anti-Forgetting in Continual Learning of LLM via General Samples Replay
Yunan Zhang, Shuoran Jiang, Mengchen Zhao, Yuefeng Li, Yang Fan, Xiangping Wu, Qingcai Chen
https://arxiv.org/abs/2508.04676

GeRe: Towards Efficient Anti-Forgetting in Continual Learning of LLM via General Samples Replay
The continual learning capability of large language models (LLMs) is crucial for advancing artificial general intelligence. However, continual fine-tuning LLMs across various domains often suffers from catastrophic forgetting, characterized by: 1) significant forgetting of their general capabilities, and 2) sharp performance declines in previously learned tasks. To simultaneously address both issues in a simple yet stable manner, we propose General Sample Replay (GeRe), a framework that use usu…