wovepaper — the arXiv observatory

The paper presents a large, controlled empirical study of tool use in web agents, examining how different tools, backbone models, and frameworks affect performance across multiple…

#web agents#tool use#empirical study#benchmark evaluation

cs.CV2023

Improve Video Representation with Temporal Adversarial Augmentation

Jinhao Duan, Quanfu Fan, Hao Cheng +2

cs.LG2023

LaplaceConfidence: a Graph-based Approach for Learning with Noisy Labels

Mingcai Chen, Yuntao Du, Wei Tang +4

cs.CL2020

The Microsoft Toolkit of Multi-Task Deep Neural Networks for Natural Language Understanding

Xiaodong Liu, Yu Wang, Jianshu Ji +8

cs.CL2023

INSCIT: Information-Seeking Conversations with Mixed-Initiative Interactions

Zeqiu Wu, Ryu Parish, Hao Cheng +4

physics.ins-det2020

Intrinsic radiation background of LaBr $_{3}$ (Ce) detector via coincidence measurements and simulations

Hao Cheng, Bao-Hua Sun, Li-Hua Zhu +5

cs.CL2023

Self-Verification Improves Few-Shot Clinical Information Extraction

Zelalem Gero, Chandan Singh, Hao Cheng +4

cond-mat.str-el2026

Twisting-operator approach to identifying gaplessness in SU( $N$ ) fermionic systems

Hao Cheng, Hang Su, Yuan Yao

cs.LG2026

Self-Boosting Vision-Language Models with Noisy Student On-Policy Self-Distillation

Shuai Wang, Daoan Zhang, Zhe Tang +2

cs.CV2023

READ: Aggregating Reconstruction Error into Out-of-distribution Detection

Wenyu Jiang, Yuxin Ge, Hao Cheng +3

cs.RO2023

RGBlimp: Robotic Gliding Blimp -- Design, Modeling, Development, and Aerodynamics Analysis

Hao Cheng, Zeyu Sha, Yongjian Zhu +1

cs.CV2025

Gaussian On-the-Fly Splatting: A Progressive Framework for Robust Near Real-Time 3DGS Optimization

Yiwei Xu, Yifei Yu, Wentian Gan +4

cs.RO2026

RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation

Shihan Wu, Xuecheng Liu, Shaoxuan Xie +81

cs.CL2025

SAS: Simulated Attention Score

Chuanyang Zheng, Jiankai Sun, Yihang Gao +12

cs.GT2023

Exploring Leximin Principle for Fair Core-Selecting Combinatorial Auctions: Payment Rule Design and Implementation

Hao Cheng, Shufeng Kong, Yanchen Deng +4

nucl-ex2021

Measurements of $^{160}$ Dy( $p, I^^{3}$ ) at energies relevant for astrophysical $I^^{3}$ process

Hao Cheng, Bao-Hua Sun, Li-Hua Zhu +20

cs.LG2026

TRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents

Leitian Tao, Baolin Peng, Wenlin Yao +5

The paper proposes TRACE, a turn-level reward assignment method that estimates credit for each tool-call in long-horizon agents using log‑ratio state values and temporal‑difference…

#credit assignment#long-horizon agents#tool use#dense rewards

cs.LG2021

Learning with Instance-Dependent Label Noise: A Sample Sieve Approach

Hao Cheng, Zhaowei Zhu, Xingyu Li +3

cs.CL2025

Dyna-Mind: Learning to Simulate from Experience for Better AI Agents

Xiao Yu, Baolin Peng, Michel Galley +6

cs.LG2025

Reinforcement Learning for Reasoning in Large Language Models with One Training Example

Yiping Wang, Qing Yang, Zhiyuan Zeng +11

cs.AI2023

TrainerAgent: Customizable and Efficient Model Training through LLM-Powered Multi-Agent System

Haoyuan Li, Hao Jiang, Tianke Zhang +6

cs.AI2024

ReasonAgain: Using Extractable Symbolic Programs to Evaluate Mathematical Reasoning

Xiaodong Yu, Ben Zhou, Hao Cheng +1

cs.CV2023

An End-to-End Framework of Road User Detection, Tracking, and Prediction from Monocular Images

Hao Cheng, Mengmeng Liu, Lin Chen

cs.CV2020

One for More: Selecting Generalizable Samples for Generalizable ReID Model

Enwei Zhang, Xinyang Jiang, Hao Cheng +7

cs.CV2021

An Empirical Study and Analysis on Open-Set Semi-Supervised Learning

Huixiang Luo, Hao Cheng, Fanxu Meng +4

cs.RO2023

Tracing the Influence of Predecessors on Trajectory Prediction

Mengmeng Liu, Hao Cheng, Michael Ying Yang

cs.AI2026

Synthetic Computers at Scale for Long-Horizon Productivity Simulation

Tao Ge, Baolin Peng, Hao Cheng +1

cs.CL2023

Optimizing Bi-Encoder for Named Entity Recognition via Contrastive Learning

Sheng Zhang, Hao Cheng, Jianfeng Gao +1

cs.CL2022

Human Parity on CommonsenseQA: Augmenting Self-Attention with External Attention

Yichong Xu, Chenguang Zhu, Shuohang Wang +7

cs.CL2025

MPL: Multiple Programming Languages with Large Language Models for Information Extraction

Bo Li, Gexiang Fang, Wei Ye +4

Publications (250)

Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving

Exploring Dynamic Context for Multi-path Trajectory Prediction

Data-Driven Dynamics Modeling of Miniature Robotic Blimps Using Neural ODEs With Parameter Auto-Tuning

Chain-of-Skills: A Configurable Model for Open-domain Question Answering

Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations

Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior

AdaMR: Adaptable Molecular Representation for Unified Pre-training Strategy

A Novel Multi-Task Model Imitating Dermatologists for Accurate Differential Diagnosis of Skin Diseases in Clinical Images

Learning Latent Local Conversation Modes for Predicting Community Endorsement in Online Discussions

Exploring Typographic Visual Prompts Injection Threats in Cross-Modality Generation Models

ACT-Diffusion: Efficient Adversarial Consistency Training for One-step Diffusion Models

Understanding Driver Cognition and Decision-Making Behaviors in High-Risk Scenarios: A Drift Diffusion Perspective

TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention

Occupancy World Model for Robots

LEDGER: Scaling Agentic Document Editing with Dependency-aware Graph Retrieval

Determining the acoustoelastic effect of longitudinal waves propagating inclined to principal stress directions in concrete: theory and experimental validation

ForceFormer: Exploring Social Force and Transformer for Pedestrian Trajectory Prediction

Knowledge-Rich Self-Supervision for Biomedical Entity Linking

Jailbreak-AudioBench: In-Depth Evaluation and Analysis of Jailbreak Threats for Large Audio Language Models

ThetaEvolve: Test-time Learning on Open Problems

Reassessing Layer Pruning in LLMs: New Insights and Methods

T-Graph: Enhancing Sparse-view Camera Pose Estimation by Pairwise Translation Graph

LinguaSim: Interactive Multi-Vehicle Testing Scenario Generation via Natural Language Instruction Based on Large Language Models

CollabLLM: From Passive Responders to Active Collaborators

Leopard: Towards High Throughput-Preserving BFT for Large-scale Systems

Disturbance-Aware Flight Control of Robotic Gliding Blimp via Moving Mass Actuation

Token Cleaning: Fine-Grained Data Selection for LLM Supervised Fine-Tuning

Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Composition

Orchard: An Open-Source Agentic Modeling Framework

Noise-Resilient Point-wise Anomaly Detection in Time Series Using Weak Segment Labels

Mamba as Decision Maker: Exploring Multi-scale Sequence Modeling in Offline Reinforcement Learning

Domain-Specific Pretraining for Vertical Search: Case Study on Biomedical Literature

Disentangled Feature Representation for Few-shot Image Classification

Boosting Private Domain Understanding of Efficient MLLMs: A Tuning-free, Adaptive, Universal Prompt Optimization Framework

Determining diffusivity and dissipation of diffuse body waves in concrete with multiple planar boundaries

Spiking Neural Network as Adaptive Event Stream Slicer

RoIPoly: Vectorized Building Outline Extraction Using Vertex and Logit Embeddings

Transfer Attack for Bad and Good: Explain and Boost Adversarial Transferability across Multimodal Large Language Models

UNIVERSE: Unified Video Action Models for Autonomous Driving with Flexible Mask-Modulated Modality Generation

Controllable Traffic Simulation through LLM-Guided Hierarchical Reasoning and Refinement

LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy

Many could be better than all: A novel instance-oriented algorithm for Multi-modal Multi-label problem

Evaluating the Efficacy of Length-Controllable Machine Translation

Interaction Detection Between Vehicles and Vulnerable Road Users: A Deep Generative Approach with Attention

AgenticMath: Enhancing LLM Reasoning via Agentic-based Math Data Generation

A Foundation Model for Spatial Proteomics

Mitigating Memorization of Noisy Labels via Regularization between Representations

RMNet: Equivalently Removing Residual Connection from Networks

Pre-training, Fine-tuning and Re-ranking: A Three-Stage Framework for Legal Question Answering

RBFormer: Improve Adversarial Robustness of Transformer by Robust Bias

On the geometry of the second Lagrange spectra

Spiking Denoising Diffusion Probabilistic Models

CRAFT: Counterfactual-to-Interactive Reinforcement Fine-Tuning for Driving Policies

Augmenting Language Models with Long-Term Memory

Defending against Backdoor Attack on Deep Neural Networks

Open-domain Question Answering via Chain of Reasoning over Heterogeneous Knowledge

LiPS: Large-Scale Humanoid Robot Reinforcement Learning with Parallel-Series Structures

MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts

Thermal Stress Analysis of the LNG Corrugated Cryogenic Hose During Gas Pre-Cooling Process

Variational Disentanglement for Domain Generalization

JUMBO: Fully Asynchronous BFT Consensus Made Truly Scalable

On Memory Construction and Retrieval for Personalized Conversational Agents

Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing

D2E-An Autonomous Decision-making Dataset involving Driver States and Human Evaluation

Efficient Encoder-Decoder Transformer Decoding for Decomposable Tasks

DyFADet: Dynamic Feature Aggregation for Temporal Action Detection

LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents

Pre-training Transformers for Knowledge Graph Completion

Unsupervised Learning of Hierarchical Conversation Structure

ChartM3: A Multi-Stage Code-Driven Pipeline for Constructing Multi-Dimensional and Multi-Step Visual Reasoning Data in Chart Comprehension

The Tool Illusion: Rethinking Tool Use in Web Agents

Improve Video Representation with Temporal Adversarial Augmentation

LaplaceConfidence: a Graph-based Approach for Learning with Noisy Labels

The Microsoft Toolkit of Multi-Task Deep Neural Networks for Natural Language Understanding

INSCIT: Information-Seeking Conversations with Mixed-Initiative Interactions

Intrinsic radiation background of LaBr3​(Ce) detector via coincidence measurements and simulations

Self-Verification Improves Few-Shot Clinical Information Extraction

Twisting-operator approach to identifying gaplessness in SU(N) fermionic systems

Self-Boosting Vision-Language Models with Noisy Student On-Policy Self-Distillation

ChartM $^{3}$ : A Multi-Stage Code-Driven Pipeline for Constructing Multi-Dimensional and Multi-Step Visual Reasoning Data in Chart Comprehension

Intrinsic radiation background of LaBr $_{3}$ (Ce) detector via coincidence measurements and simulations

Twisting-operator approach to identifying gaplessness in SU( $N$ ) fermionic systems

Measurements of $^{160}$ Dy( $p, I^^{3}$ ) at energies relevant for astrophysical $I^^{3}$ process