Research Paper Cockpit

Daily Digest - 2026-11-03

Papers first seen in this daily snapshot.

Daily Archives

Quick jump into generated daily digests.

Research Workflow

Latest digest: 2026-11-17.

Papers

66 visible entries

arxiv Score 31.8

Can Large Language Model Agents Learn by Watching Tutorial Videos?

2026-11-01 · Haoyu Zhao, Zafeirios Fountas, Jun Wang

Research Track A · General AI

Humans continuously acquire knowledge from carefully crafted materials such as books and tutorial videos, carrying that understanding into future tasks. Millions of such instructional videos are uploaded online every day, yet LLM agents remain unable to exploit this ever-expanding wealth of human teaching. Their capabi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 29.5

BISECT: Bilevel Spectral Decomposition for Low-Rank Continual Learning

2026-11-02 · Tom Saliencro, Wen-Hsuan Liu, Rohan Desai, Amara Okonkwo, Maya Lindqvist

Research Track A

Exemplar-free continual learning on pre-trained Vision Transformers increasingly relies on Low-Rank Adaptation (LoRA), yet prevailing methods govern the low-rank update space through strict subspace isolation. By projecting each incoming task into the approximate null space of historical token activations, isolation-ba…

Review
pending
Role
unreviewed
Read
now
arxiv Score 29.5

RASP: Read-All, Sparse-Plasticity Continual Learning in a Stationary-Size LoRA

2026-11-02 · Wen-Hsuan Liu, Amara Okonkwo, Tom Saliencro, Rohan Desai, Maya Lindqvist

Research Track A

Parameter-efficient continual learning on pre-trained Vision Transformers largely relies on a costly premise: preventing catastrophic forgetting requires either spawning a dedicated LoRA branch per session---causing adapter parameters to scale linearly with task horizon---or applying sparse Mixture-of-Experts gating sy…

Review
pending
Role
unreviewed
Read
now
arxiv Score 28.0

HQNN-CL: Regime-Aware Hybrid Quantum Neural Network with Continual Learning for Financial Forecasting

2026-11-01 · Muhammad Atif Saeed, Akhtar Jamil, Subhan Ullah

Research Track A

Financial time-series forecasting is still difficult owing to the non-stationary nature of the market, regime changes, and catastrophic forgetting, which makes conventional deep learning models hard to adapt in the long term. To achieve uncertainty-aware robust financial forecasting in the face of changing market condi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 24.0

WebHarbor: Docking Real Websites for Evolving GUI Agent Environments

2026-11-02 · Zhaoyang Wang, Qianhui Wu, Boyu Gou, Demi Wang, Jiacheng Zhang, Peichun Hua, Qize Li, Shi Qiu, Yifan Jiang, Yifan Zhang, Zexu Jin, Baolin Peng, Bolin Shen, Chetan Bansal, Dehan Kong, Deren Lei, Dongshen Peng, Fengqing Jiang, Hao Cheng, Haoqing Wang, Jiaming Shan, Jicheng Wang, Jingguang Zhou, Jixuan He, Kaide Zeng, Kangyu Zhu, Lama Moukheiber, Mufan Qiu, Peng Xia, Qihan Sun, Qingwei Lin, Qiushi Sun, Ruichen Zhang, Shuqing Luo, Si Qin, Sicheng Fan, Tao Xiong, Wenlin Yao, Xinyu Zhang, Xinyue Zeng, Xuanrui Li, Xuchao Zhang, Yian Wang, Yichao Liu, Yiming Liang, Yiyang Zhou, Yizhu Jiao, Yuetai Li, Yurun Chen, Yuxuan Zhang, Zheyuan Yang, Zhiyuan Yao, Zhongyang Li, Zihao Li, Zili Wang, Jianfeng Gao, Huaxiu Yao

Research Track B · General AI

GUI web agents require realistic and interactive environments that remain stable during both evaluation and training. Live website evaluation suffers from changing content, access restrictions, and limited coverage of authenticated workflows, while existing synthetic environments for training often lack functional dept…

Review
pending
Role
unreviewed
Read
now
huggingface Score 22.0

DR Pixel: Page-Native Deep Research with Vision-Language Models

2026-11-02 · Tiansheng Hu, Yichuan Wang, James V. Talwar, Nicholas Tomlin, Chen Zhao, Sewon Min

General AI

Complex documents such as scientific papers contain rich information in tables, figures, and equations; however, existing deep research (DR) agents typically reduce papers to OCR-extracted text. We instead ask whether vision-language models (VLMs) can enable page-native deep research: directly retrieving and reasoning …

Review
pending
Role
unreviewed
Read
now
arxiv Score 19.5

From Local Skills to Long-Horizon Web Tasks: Progressive Skill Exploration for LLM Web Agents

2026-11-02 · Xueyang Feng, Xiaohe Bo, Xu Chen, Quanyu Dai, Feng Wen

Research Track B · General AI

Large language models (LLMs) are driving autonomous agents toward real-world environments, with the Web emerging as one of the most representative interaction scenarios. However, training Web agents with naive exploration remains challenging: successful trajectories are rare, and the sparse terminal rewards make it dif…

Review
pending
Role
unreviewed
Read
now
arxiv Score 18.8

Reviso: Not All Visual Evidence Deserves Equal Attention in Video Captioning

2026-11-02 · Tengfei Liu, Yang Shi, Yuran Wang, Yuqi Tang, Yizhuo Jia, Zhuoran Zhang, Qixun Wang, Shujie Li, Xuanyu Zhu, Xinyu Liu, Minghui Guo, Shixuan Liu, Haotian Wang, Yuanxing Zhang, Pengfei Wan

General AI

Multimodal large language models (MLLMs) can generate detailed video captions, yet they may overemphasize minor details while overlooking essential context and key events. We argue that not all visual evidence deserves equal attention, as the importance of visual information depends on the content of each video. Motiva…

Review
pending
Role
unreviewed
Read
now
arxiv Score 18.0

Consolidating Reasoning with Test-Time Training

2026-11-01 · Haotian Wu, Zeming Chen, Hao Zhao, Antoine Bosselut

Research Track A · General AI

Effective multi-agent collaboration requires finalizing solutions from reasoning that is distributed across the individual attempts of many agents. However, useful intermediate reasoning can remain unexploited when collaboration is limited to voting on conclusions reached among individual agents, or aggregating long, n…

Review
pending
Role
unreviewed
Read
now
arxiv Score 17.8

MultiInpaintCOCO: Fine-Grained Visual Understanding in Multiple Languages

2026-11-02 · Andrei-Alexandru Manea, Jindřich Libovický, Philipp J. Rösch

General AI

Previous research shows that vision-language (VL) models struggle to understand fine-grained image details and tend to focus on the overall layout. Moreover, any weakness in English tends to be further amplified in the multilingual setting. To study that, we extend InpaintCOCO, an English dataset focused on fine-graine…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.8

R$^2$-Omni: Reducing Structural and Precision Redundancy in Omni-LLMs for Long-Form Generation

2026-11-02 · Bolong Feng, Pengfei Zhou, Zhenglin Wan, Wei Wang, Wangbo Zhao

Research Track A · General AI

Omni-modal large language models (Omni-LLMs) typically process large numbers of multimodal tokens from audio-visual inputs, incurring substantial computational costs and a large memory footprint for the key-value (KV) cache. Aggressive token pruning can alleviate this problem while maintaining competitive performance o…

Review
pending
Role
unreviewed
Read
now
huggingface Score 15.0

Advances in Agentic Computer Graphics

2026-11-01 · Hengyu Liu, Jihyeon Je, Ruiqi Wang, Mingxuan Wu, Haven Feng, Qixing Huang, Georgios Pavlakos, Hao Zhang, Jiajun Wu, Maneesh Agrawala, Leonidas Guibas, R. Kenny Jones, Shangzhe Wu, Guandao Yang

General AI

Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progr…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.0

Do agents need to sleep?

2026-11-01 · Ido Haber, Giulio Tononi

Research Track A · General AI

LLM-based agents, which increasingly operate persistently across sessions and accumulate memory through interaction, already exhibit the familiar symptoms of unchecked plasticity: semantic drift, degraded retrieval precision, propagation of stored errors, and saturation of memory capacity. Drawing on the selectionist "…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

FADE: Factored Ability--Difficulty Filtering for Online RLVR Scheduling

2026-11-02 · Tomas Duncun, Jonah Ellison, Mateo Ruiz, Hanae Fujita, Ingrid Solberg

General AI

Group-based reinforcement learning with verifiable rewards (RLVR) learns only from prompts whose sampled completions disagree: all-pass and all-fail groups produce zero relative advantage and burn generation budget without moving the policy. Adaptive schedulers try to steer rollouts toward informative prompts, yet they…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

Not All Reasoning Generalizes: A Case Study on Learning Generalizable Reasoning Strategies in RLVR

2026-11-02 · Ziqi Pang, Yu-Xiong Wang

General AI

Reinforcement learning with verifiable rewards (RLVR) can enable generalizable improvement for large language model (LLM) reasoning. This paper studies how to incentivize such generalization via reasoning strategies. We study this question with a controllable equation-solving framework: finding multiple inputs x for eq…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

Same Outputs, Different Scores: Auditing the Unchecked Graders of Multimodal LLM Benchmarks

2026-11-02 · Ahmad Jarjis Hasan, Jonathan Lwowski

General AI

Every multimodal-LLM multiple-choice benchmark ends with a scoring rule that converts a model's response, often free-form text rather than a clean option letter, into a right/wrong verdict, buried in its code. Unlike the dataset, this rule is chosen without justification and left unexamined, yet it sets the published n…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

TAMPAgent: Agentic Belief-Space Planning for Mobile Manipulation

2026-11-02 · Utkarsh A. Mishra, Alperen Degirmenci, Yongxin Chen, Danfei Xu, Tucker Hermans, Caelan Garrett

General AI

Open-world robotic mobile manipulation requires acting in physical environments with only partial information. While recent Large Vision-Language Model (LVLM) agents show strong open-world reasoning capabilities, existing architectures that rely on direct tool calling struggle with geometric uncertainty and thus are li…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

VeriHarm-Bench: Measuring Harmful Success in LLM Coding Agents Verifiably

2026-11-02 · Wei-Lin Chen, Zhepei Wei, Xinyu Zhu, Yu-Min Tseng, Tzu-Han Lin, Yu Meng

General AI

Large language model (LLM) coding agents increasingly operate with shell access, allowing them to write scripts, move files, alter permissions, and modify system configurations. However, an agent may achieve a user's goal while causing unintended harm, such as deleting unrelated files, weakening access controls, or cor…

Review
pending
Role
unreviewed
Read
now
huggingface Score 14.0

DAMI: Dual-Axis Memory Initialization for Enhanced Long-Context Retrieval in Gated Linear Attention

2026-10-30 · Zhangchi Feng, Harry Jiang, Yangyifei Luo, Ray Zhang, Richong Zhang

Research Track A · General AI

Linear attention scales linearly with sequence length, but its finite-state memory can limit long-context retrieval. In gated linear recurrences, memory retention depends on state decay, which is shaped jointly by gate parameters A_log and dt_bias and by input activations. In controlled Gated DeltaNet (GDN) experiments…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.8

Fuse To Align: Multimodal Representation Learning for Retrieval via Hypergraph Regularization

2026-11-02 · Anindya Nag, Ambuj Mehrish, Sebastiano Vascon

General AI

Multimodal retrieval must integrate complementary evidence from views that may be missing, noisy, or unequally informative. When a sample provides limited reliable evidence, learning from its views alone can hinder the acquisition of useful retrieval representations. Semantically related training samples may provide co…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.8

Strategist: Evolving Reasoning with Agents

2026-11-02 · Taiming Lu, Qi Xu, Zhigang Hua, Zhuang Liu

General AI

A model is only half of an AI agent system. The other half is the inference strategy: the reasoning procedure that turns model calls into a response. Today that second half is written by hand, and the experience of building one strategy never carries to the next. We introduce Strategist, a meta-agent that evolves infer…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.8

VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations

2026-11-02 · Avni Mittal, Avinash Anand, Dikshant Kukreja, Erik Cambria, Rajiv Ratn Shah, Sandipan Dandapat

General AI

Organisations increasingly deploy large language model agents that must follow organisational values. We check whether activation steering, which adds vectors to a model's internal activations during generation without changing its weights, can improve compliance beyond prompting alone. Since most scenarios involve sev…

Review
pending
Role
unreviewed
Read
now
huggingface Score 13.0

SIMIT: Self-Improving Vision-Language Models via Imagination at Test-Time

2026-10-11 · Mehmet Onurcan Kaya, Desmond Elliott, Dim P. Papadopoulos

Research Track A · General AI

We introduce SIMIT, a test-time self-improvement framework in which a vision-language model creates its own query-specific training data before answering an unlabeled test query. Our goal is to improve performance on a given target test set without annotations or external models. To pursue this goal, some test-time met…

Review
pending
Role
unreviewed
Read
now
huggingface Score 13.0

Multi-Agent Relay Thinking

2026-10-27 · Xu Liu, Jungang Qiao, Guikun Chen, Xiaojuan Qi, Wenguan Wang

General AI

Existing multi-agent collaboration (MAC) methods have primarily explored a group thinking collaboration topology, where agents independently generate solutions and aggregate their outputs. In this work, we investigate an alternative relay thinking topology, where reasoning ownership is dynamically transferred across ag…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.0

Replay for LLM Continual Learning with Multi-Teacher On-Policy Distillation

2026-11-02 · Yiqiao Qiu

Research Track A · General AI

Multi-teacher on-policy distillation (MOPD) distills domain experts into one student but needs every expert at once. When domains arrive one at a time and earlier experts are gone, the previous-stage student can teach replayed prompts of earlier domains inside every stage. Some pipelines instead store every stage snaps…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

DiabAgent: Toward Agentic Physiological Control

2026-11-02 · Chengxi Li, Shaoxuan Li, Deyao Hong, Qingle Liu, Yi Huang, Puheng Bian, Minzhi Xu, Lanyue Zhang, Mingze Sun, Xin Li, Calvin Xiao, Shengyuan Wang, Huan-ang Gao, Jiazheng Xu, Kaisen Yang, Hongning Wang

General AI

As pretraining scales, medical language models have advanced rapidly in clinical question answering. Their ability to achieve effective physiological control through sustained intervention still lacks systematic quantitative evaluation. Type 1 diabetes provides a concrete setting: impaired insulin secretion creates dep…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

From Natural Language to Exact Cover: A Neuro-Symbolic Approach to Zebra Puzzles

2026-11-02 · Paulius Skaisgiris, Thomas Pammer, Veronika Semmelrock, Mykyta Ielanskyi, Maximilian Heisinger, Erich Kobler

General AI

Chain-of-Thought (CoT) generation has substantially improved the performance of Large Language Models (LLMs) on complex reasoning tasks, including code generation, data analysis, and exam-style question answering. Despite these advances, purely neural LLMs continue to struggle with elementary logical reasoning problems…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

SMILE: A Structured Multi-Agent Clinical Reasoning for Liver Ultrasound Understanding

2026-11-02 · Haris Ghafoor, Minjee Seo, Seonaeng cho, Keungmo Yang, Jaejun Lee, Kyungho Yoon

General AI

Automated liver ultrasound report generation improves the diagnosis of liver diseases by producing clinically structured descriptions of anatomical and disease-related findings. However, current generative vision-language models (VLMs) often rely on a single reasoning pathway that maps ultrasound appearance directly to…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.0

Cross-Primitive Consistency in Jev

2026-10-28 · Xu Liu, Guikun Chen, Kanzhi Wu, Wenguan Wang

General AI

Jev enables efficient decision-making with large language models (LLMs) through typed querying primitives, allowing the same semantic state to be expressed in different formats, such as binary and multiple-choice queries. However, the consistency of decisions across primitives has not been systematically evaluated. Thi…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.0

Where Do Vision Language Models Fail at 3D Reasoning?

2026-10-29 · Khai Le-Duc, Nguyen Kim Hai Bui, An Ngo, Nam D. Hoang, Yueqi Song, Jiarui Liu, Thai Quoc Hoang, Fang Wu, Chris Ngo, Terry Jingchen Zhang, Mona T. Diab, Ran Xu, Silvio Savarese, Xiao-Ming Wu

General AI

Vision-language models can get individual judgments right while struggling to recover and consistently use the exact properties of a 3D object. To characterize this gap, we introduce BedBench, a controlled benchmark of thirteen procedurally generated tasks spanning individual judgments and composite answers. Across twe…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.0

PatchScout: Thematic Web Data Collection via Information Foraging

2026-10-30 · Michael West, Eduard Dragut

Research Track B · General AI

Researchers often need web corpora that are not answers to a single query, but reusable collections of documents spanning many decentralized sources. We refer to this problem as thematic web data collection: assembling semantically relevant documents that share a common theme but are scattered across heterogeneous and …

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.0

Lab2Sim: Turning Real-World Experiment Videos into Interactive Virtual Experiments

2026-11-02 · Xusen Hei, Jiali Chen, Jinyu Yang, Yi Cai

General AI

Science experiment videos demonstrate how experiments are performed, but offer no direct interaction for learners. Turning them into virtual experiments requires not only reproducing the observed apparatus and state changes, but also enforcing the preconditions governing each operation. We introduce Lab2Sim, a benchmar…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Benchmark of THReD: Evaluating an LLM Agent and Its Preprocessing Tools for Turning Electronic Health Records into AI-Ready Datasets

2026-11-02 · Javad Mohammad Alizadeh, Huanmei Wu

Research Track A · General AI

LLM agents are increasingly built into data-analysis software, yet little evidence shows how reliably they preprocess clinical data or whether the tools they invoke give correct, reproducible output. We report a benchmark of THReD, a web-based system that turns electronic health record (EHR) extracts into AI-ready data…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Exploring Proximity and Tactile Sensing for Robotic Manipulation

2026-11-02 · Eric T. Chang, Bianca Aumann, Zhanpeng He, Mohamed Elnoor, Thomas Liang, Matei Ciocarlie, Nima Fazeli, Joshua Migdal, Tye Brady, Taskin Padir

General AI

Pre-touch sensing, or sensing information about an object before contact, is an attractive modality for robot hands as it can allow a robot to anticipate and respond to a contact before it happens. In this work, we introduce ToFTouch, a multimodal tactile sensor combining three multizone (4$\times$4 resolution) time-of…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Four Bridges: Measuring Deception Propensity in a Multi-Agent Game

2026-11-02 · Raphael Casey, Michael Simeone, Alberto Tosato, Florian Ernst, Kemal El Moujahid, James Tamplin, Tommaso Tosato

General AI

AI agents are increasingly deployed in settings where they interact and coordinate with other agents, yet honesty is mostly measured in single-turn, single-agent benchmarks that pressure the model to deceive. We study deception that emerges in multi-agent interaction without any instruction to deceive. We introduce Kra…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Tracing the Scientific Discovery Capacity of LLMs to Their Training Data

2026-11-02 · Arda Uzunoglu, Mahler Revsine, Michael Schatz, Benjamin Van Durme, Mehrdad Farajtabar, Daniel Khashabi, Eric Nalisnick

General AI

Large language models (LLMs) show growing promise for scientific discovery, yet basic questions remain about what enables reliable discovery. How can we trace the scientific discovery capacity of LLMs to their training data? We study this question as one of missing probability mass and develop a theoretical account in …

Review
pending
Role
unreviewed
Read
now
arxiv Score 9.8

EgoActionBench: Benchmarking Dense Action Captioning for Egocentric Human Videos

2026-11-02 · Yujiao Shen, Jiahe Song, Haitian Li, Jingkang Yang, Ziwei Liu

General AI

Learning from human data offers a scalable path toward embodied intelligence, yet it requires models to accurately understand fine-grained human actions from egocentric observations. However, existing benchmarks remain limited in action atomicity, temporal precision, and hand-action association. We introduce EgoActionB…

Review
pending
Role
unreviewed
Read
now
arxiv Score 9.8

PlantVLM-Edge: Quantized Vision-Language Models for On-Device Crop Health Diagnosis

2026-11-02 · Gaurab Wagle, Bishal Adhikari, Jiajia Li, Xiaolei Huang, Dong Chen

General AI

Timely and accurate crop health diagnosis is essential for sustainable agricultural production, yet accurate diagnosis often requires expert knowledge that is not readily available. Recent vision-language models (VLMs) have demonstrated remarkable capabilities in visual understanding, reasoning, and natural-language in…

Review
pending
Role
unreviewed
Read
now
arxiv Score 9.5

LADS: Lease-Aware Delivery Scheduling of Tool Results for Language Model Agents

2026-10-31 · Yuxuan Hou, Linshan Jiang

Research Track B · General AI

Web agents complete multi-step online tasks through repeated browser actions and model calls. Their returned observations lie on the critical path to subsequent decisions. When concurrent browser actions finish in bursts, observations can accumulate on a constrained return path to a shared inference service, while wait…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.5

When 200 OK Is Not an Answer: Silent Fetch Failures in Agentic Web Content Acquisition

2026-10-31 · Shuyang Zhang

Research Track B · General AI

Web agents and answer engines fetch pages at question time, and liveness checks of cited sources treat an HTTP 200 response as success. We measure how often that assumption fails and what it does to measurements built on fetched pages. We fetched 10,307 URLs in three days with a polite static client and a rendered brow…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.8

Evaluating Depth-Augmented Visual Features for Cross-Dataset Lunar Traversability Estimation

2026-11-02 · Jan Baturo, Bartosz Ptak

General AI

Autonomous mobility is essential for sustained lunar surface operations, as communication delays and line-of-sight occlusions limit continuous teleoperation. Although recent traversability estimation methods increasingly rely on deep learning, the domain gap between terrestrial analogue environments and real lunar terr…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.8

Pretraining Data Scale Reverses the Muon-AdamW Fine-Tuning Gap

2026-11-02 · Catherine Li, Jingchu Gai, Andrej Risteski, Jacob Mitchell Springer, Aditi Raghunathan

General AI

Muon has recently emerged as a promising optimizer for pretraining large language models, offering a speedup over AdamW in minimizing pretraining loss. However, it remains unclear whether Muon's advantage persists after fine-tuning. In this paper, we study the downstream behavior of Muon and AdamW and uncover a surpris…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.3

Spacetime Emergence in the Memory of a Learning Agent

2026-11-02 · Philip A. LeMaitre, T. Rick Perche

General AI

We show that the causal structure of spacetime can emerge within the memory of a learning agent performing a quantum information task. We train a model-based reinforcement-learning agent to maximize the quantum mutual information between localized quantum probes interacting with a quantum field in Minkowski spacetime, …

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.0

WorldFlow: Joint Video and 3D Motion Prediction via Large-Scale Pretraining

2026-10-29 · Yuhao Lu, Jiongxu Chen, Hexiang Wei, Jiajin Wu, Juelin Li, Mingjian Liang, Fan Yang, Zhuowei Liang, Zihao Liu, Zhennan Wang, Sheng Wang, Lin Shao

General AI

Predicting future physical states stands as a central challenge in embodied intelligence. Existing video world models leave 3D dynamics implicit, action-based formulations remain tied to specific robot configurations, and 3D motion prediction models use flow-only supervision, providing a limited learning signal for phy…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.0

SWE-Trail: History-Guided Trajectory Collection for Coding Agent Training

2026-11-01 · Ziming Li, Lu Yu, Jiatan Huang, Qing Cui, Jun Zhou, Chuxu Zhang

General AI

Coding agents solve repository issues through long sequences of exploration, editing, and validation. These trajectories are increasingly used as post-training data, so how they are collected shapes what the trained model learns. A common approach is to sample multiple trajectories independently for each issue and then…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.0

LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents

2026-11-02 · Sreehari Sankar, Debargha Ganguly, Vikash Singh, Mohsen Hariri, Weicong Chen, Yanyan Zhang, Shouren Wang, Wang Yang, Chaoda Song, Alexander Nemecek, Osama Zafar, Idan Davidovich, Yuting Shao, Kai Ye, Biyao Zhang, Xinpeng Li, Joe Stubbs, Erman Ayday, Jing Ma, Vipin Chaudhary

General AI

Tool-using language-model agents often execute actions that their domain policies forbid. We present LeanGuard, a runtime that checks every proposed tool call against a policy written in Lean4 before the call reaches the environment. Policies combine state predicates with scoped past-time temporal operators over the ex…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.8

Sensitivity Inflation from Cross-Record Precision Routing in Private Quantized Fine-Tuning

2026-11-01 · Annan Zhang, Shengjie Song, Wenshuai Yao, Lihao Zhang, Tianmiao

General AI

Private fine-tuning of large language models increasingly pairs DP-SGD with low-rank adapters on frozen, quantized backbones, and every such stack inherits one unstated premise: frozen layers keep each record's clipped gradient a function of that record alone. We show that LLM.int8 breaks this premise. It routes outlie…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 7.0

When Morphs Look Better: Face Image Quality as a Training-Free Signal for Morphing Attack Detection

2026-10-15 · Hao Peng, Fu Bo, Mahmoud Elshirbeny, Guray Ozgur

General AI

Face image quality assessment (FIQA) estimates the utility of a face image for automated face recognition and has been proposed as a morphing attack detection (MAD) signal that requires no morph-specific training. We evaluate 17 FIQA methods from all current FIQA categories on 20 morphing attack sets from two datasets,…

Review
pending
Role
unreviewed
Read
later
huggingface Score 7.0

When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models?

2026-11-01 · Anton Hristov, Aleksandar Slavov, Vasil Vasilev

General AI

Prediction errors from video world models are often treated as physical surprise, but they can also reflect appearance, static content, motion, or ordinary prediction difficulty. We present an evaluation framework that separates what frozen representations contain, how they organize appearance and physical differences,…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

ECGJev: Question-Conditioned Option Scoring for 12-Lead ECG Interpretation

2026-11-01 · Zhuoyuan Li, Xidong Wu, Hanyu Cui, Jiansong Gao, Rongrong Fu, Mingke Yan, Haotian Tang

General AI

Clinical interpretation of a 12-lead electrocardiogram (ECG) often requires answering targeted diagnostic questions with clear, quantifiable decisions. We introduce ECGJev, the first ECG framework built on a decision decoder, a pretrained language model trained to return a probability for every listed option. A questio…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

ArabGeoText: Tool-Augmented Vision-Language Models for Arabic Geospatial Text Spotting

2026-11-02 · Yakoub Bazi, Sarah Alsaleh, Mohamad M. Al Rahhal, Mohamed A. Mekhtiche, Mansour Zuair

General AI

Geospatial text spotting involves locating and transcribing geographic labels, such as road names, district names, and landmarks, in raster maps. These labels are often small, densely arranged, and arbitrarily oriented. Although Arabic text recognition has been studied extensively in documents and natural scenes, Arabi…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

CL3DOD: Understanding the Stability-Plasticity Dilemma in Outdoor Class-Incremental 3D Object Detection

2026-11-02 · Muhammad Ahmed Ullah Khan, Khalil Ahmed, Didier Stricker, Muhammad Zeshan Afzal

General AI

Class-incremental learning (CIL) for outdoor 3D object detection remains largely underexplored. Existing outdoor studies are limited to single-benchmark evaluations, with SDCoT++ considering KITTI and CL2OD considering nuScenes under different incremental settings. This makes it difficult to assess whether observed con…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

Pixel-UMI: Learning Environment-Aware Whole-Body Manipulation from Sensorized Human Demonstrations

2026-11-02 · Jason Jingzhou Liu, Tony Tao, Deepak Pathak, Ruslan Salakhutdinov

General AI

Sensorized human demonstrations offer scalable data for mobile manipulation, but embodiment differences complicate policy learning. We present Pixel-UMI, a pipeline for learning environment-aware whole-body policies from human data. We first introduce Environment-Aware Retargeting (EAR), which combines reconstructed sc…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

Pixel: A Co-Designed Full-Stack Robot System for Learning General-Purpose Mobile Manipulation

2026-11-02 · Tony Tao, Jason Jingzhou Liu, Ruslan Salakhutdinov, Deepak Pathak

General AI

General-purpose mobile manipulation requires a combination of capable robot hardware, scalable demonstration collection, expressive whole-body control, and effective policy learning. Yet these components are often developed independently, creating mismatches that limit the capabilities of the complete system. We presen…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 6.5

Can Coding Agents Sustain Continual Evolution without Forgetting?

2026-11-02 · Shangziqi Zhao, Yitong Zhang, Jia Li

General AI

Recently, harness self-evolution has emerged as an important paradigm for improving agent performance. Prior studies show that evolving the agent harness can effectively enhance agent capabilities without modifying the foundation model's parameters. We find that existing approaches largely study agent evolution on a si…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

CARE: Grounded Recursive Self-Improvement for Production Customer-Service Agents

2026-11-02 · Yanbei Jiang, Chao Lei, Rena Gao, Xiaochen Zhao, Yuchen Ge, Haohang Liu, Runze Nie, Qitian Chen, Qi An, Haolin Song, Yufei Sun, Yihao Ding, Jey Han Lau, Krista A. Ehinger, Xingjun Ma, Weijia Chen, Ruiyuan Wu, Zhen Luo, Ke Xing, Yongjun Yin

General AI

Recursive self-improvement (RSI) in production agents involves transforming accumulated experience into reliable model updates while adapting to evolving business policies. We present CARE (Customer-service Agent Retrospective and Evolve), a grounded RSI framework that operationalizes this process in customer service. …

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

Comparative study of global pooling for pre-departure ship fuel consumption prediction: Advantages over per-ship and adaptation-based approaches

2026-11-02 · Heungjun Im, Bongmin Cha, Agung Nugraha, Donghyun Kim, Seongphil Kang, Minwook Choi, Jeongmin Park, Jihwan Lee

General AI

Most fuel oil consumption (FOC) models for voyage planning are per-ship models and thus generalize poorly beyond each vessel's records. Transfer and federated learning have been reported to perform better, but under different protocols that often use inputs unavailable before departure or random data splits. It therefo…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

Isekai3D SDG: A Platform for Diverse Synthetic Gameplay Visual Data

2026-11-02 · Rafal Karp, Gene Liu, Shih-Hsin Li, Eric Courtemanche, Shengze Wang, Josef Spjut, David Pankratz, Mischa Kolbe, Michael Stengel, Sven Elflein, Adam Grygielski, Liwen Wu, Anjul Patney

General AI

We present Isekai3D Synthetic Data Generation (SDG), a platform for simulating and capturing synthetic 3D gameplay for research in vision and world models, neural graphics, gameplay agents, and other applications. The platform combines reusable environment and character assets with Unreal Engine plugins in a two-stage …

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

Neural Actor--Critic Learning with Minimax Critics and Reflected Langevin Sampling

2026-11-02 · Kunqi Huang, Lican Kang, Hongrui Shu, Jincheng Ying

General AI

Actor--critic methods rely on critic gradients to improve continuous-action policies, yet finite-sample guarantees must control both value and gradient errors and their propagation through the actor. We develop a neural actor-critic procedure that combines a regularized minimax Bellman estimator over sigmoid networks w…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

One Image Is Enough: Self-Generated Supervision for Personalizing Human Video Diffusion at Test Time

2026-11-02 · Sadman Rakib Pinon, Adamo Orsini, Yanis Falaki, Linlian Jiang, Jacob Kingi, Yang Wang, Xinxin Zuo

General AI

Pose-driven human video diffusion animates a person from a single reference image and a target pose sequence, yet it struggles to preserve identity, rendering the face and garments unfaithfully as the pose departs from it. Personalization is the standard remedy, but its supervision is missing: one image fixes identity …

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

Should We Combine Our Data? Algorithms and Limits for Privately Assessing Data Sharing

2026-11-02 · Mahdi Haghifam, Kumar Kshitij Patel, Avrim Blum, Giulia Fanti, Juba Ziani

General AI

In this paper, we study when two parties should pool their data to train a machine learning model. Two agents, Alice and Bob, each hold a sensitive labeled dataset and want to train a classifier from a given hypothesis class $\mathcal{H}$. Even when each party's dataset can be perfectly classified on its own, the two p…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 5.0

PhotoGuider: Composition-Aware Cropping and 360^{circ} View Framing with Explainable Photographic Assessments

2026-11-02 · Yuan-Kang Lee, Kuan-Lin Chen, Chia-Che Chang, Yu-Lun Liu

General AI

Traditional image cropping assumes the photograph already contains the composition worth keeping, and asks only where to crop. A full 360^{circ} scene offers no such guarantee: the viewing direction in a scene has to be discovered before anything can be framed. We present PhotoGuider, a measure-then-frame method that a…

Review
pending
Role
unreviewed
Read
later
huggingface Score 5.0

TaoMate-FVSR:Audio-Guided Facial Video Restoration

2026-11-02 · Xinyang Li, Zeran Xu, Qijun Gan, Chenwei Zhang, Meiguang Jin, Junfeng Ma

General AI

AI-generated videos often depict people within detailed scenes while leaving their small facial regions poorly resolved. Visual super-resolution can sharpen these regions without resolving whether the recovered lips and teeth agree with the spoken audio. We present an audio-guided facial video restoration framework tha…

Review
pending
Role
unreviewed
Read
later
arxiv Score 4.8

DINOFAS: Bonafide-Centric Representation Learning for Face Anti-Spoofing

2026-11-02 · Shakeel Muhammad Ibrahim, Muhammad Sohail Ibrahim, Muhammad Shahid Jabbar, Shujaat Khan

General AI

Face anti-spoofing (FAS) models often learn attack- and dataset-specific cues that degrade performance under unseen acquisition conditions and presentation attacks. We present DINOFAS, a DINOv3-based framework that formulates FAS as bonafide-centric representation learning. The proposed framework combines parameter-eff…

Review
pending
Role
unreviewed
Read
later
arxiv Score 4.8

LODox: A Web-Based Platform for Cultural Heritage Linked Data Discovery

2026-11-02 · Puyu Wang, Rui Zhao, Hang Dong

General AI

Recent advances in Large Language Models have transformed how data can be linked, discovered, and enhanced. LODox, Linked Open Data at Oxford, is an evolving web-based platform that embraces this transformation for cultural heritage Linked Open Data. LODox features a novel interface with intuitive card-based data prese…

Review
pending
Role
unreviewed
Read
later
arxiv Score 4.8

Unlocking the Potential of Vision Foundation Models for Multi-Task Dense Prediction via Mixture of Task Adapters

2026-11-02 · Yunzhi Zhuge, Xinzhuo Yu, Sitong Gong, Lu Zhang, Pingping Zhang, Huchuan Lu

General AI

Multi-task dense prediction seeks a single, coherent representation that supports simultaneously discriminative semantic recognition and regularized geometric/structural inference. For vision foundation models, pretrained semantic invariances conflict with the frequency and smoothness biases of dense objectives; joint …

Review
pending
Role
unreviewed
Read
later