27 papers
QREF: Improving DeepSearch Agents by Aligning Search with Reasoning via Query Refinement and Evidence Filtering
Yujia Chen, Yuhui Wang, Junzhe Wang +15
DeepSearch agents achieve strong performance on complex information-seeking tasks through iterative search and reasoning. However, they often issue intent-misaligned, under-constra…
WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution
Hang He, Linting Wang, Guanming Liu +14
Scaling interactive environments is important for training long-horizon agents through reinforcement learning. These environments often require further testing and refinement after…
LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets
Jun Zhao, Leiming Fu, Yanbo Wen +9
Evaluating agents by outcomes alone can obscure the capabilities that produce them. This problem is especially pronounced in evolving environments, where outcomes reflect a closed-…
EvoIn: Bridging Evolution and Internalization for Agent Fine-Tuning
Shihan Dou, Shaofan Liu, Zhonghang Lu +8
Recent work has explored improving agents by jointly evolving their harnesses and models, but often takes a ''potpourri'' approach that bundles together new tools, new decision-mak…
ReDraft, Don't Just Distill: Reference-Driven Revision for Continual VLLM Post-Training
Zhihao Zhang, Mingqi Wu, Qiaole Dong +15
Continual post-training of large multimodal models should add new capabilities while preserving those from pre-training, and the two goals pull in opposite directions. SFT gives ex…
SWE-Bench Pro Verified: A Reliable Benchmark for Software Engineering Agents
Pujun Zheng, Zixin Shang, Shufan Jiang +5
SWE-Bench Pro has emerged as a standard benchmark for evaluating software engineering agents on challenging repository-level tasks. However, our analysis work show that its evaluat…
Strategy Accumulation and Guided Execution for Automated LLM Fine-Tuning
Haoran Zhao, Wei Du, Dingwen Yang +7
Producing task-specific large language models requires discovering effective training strategies through experimentation. Automated fine-tuning systems have made this experimentati…
CAFE: Self-Improving Search Agents Need Co-Evolving Feedback
Boyang Liu, Senjie Jin, Peixin Wang +17
Reliable search requires more than acquiring external evidence. An agent must also recognize and recover from errors as its trajectory unfolds. In-trajectory feedback provides a me…
SPIEval: Evaluating Large Language Models as Mobile Assistants over Scattered Personal Information
Junjie Ye, Zhuohui Sheng, Shaofan Liu +12
Large language models (LLMs) are increasingly deployed as mobile assistants, where a key challenge is leveraging personal information scattered across multiple applications (apps)…
IACM-RL: Intent-Aware Context Management and Reinforcement Learning for Complex Tool Invocation under Dynamic Intent Fluctuations
Dingwei Zhu, Jiahan Li, Chengjun Pan +22
Executing long-horizon tool invocations in real-world environments is severely challenged by dynamic user intent noise. Existing methods attempt robustness via implicit history sca…
Beacon: Knowing When and How to Perform Agentic Visual Reasoning
Qixun Wang, Yang Shi, Letian Cheng +11
The fundamental goal of agentic visual reasoning is to improve the success rate of multimodal large language models (MLLMs) on complex tasks. We rethink agentic visual reasoning th…
HiSkill: Empowering LLM Agents with Hierarchical Skill Graphs
Yu Hao, Jinxuan Cai, Qi Zhang +4
Skills have become an important abstraction for enabling large language model (LLM) agents to reuse past experience in long-horizon interactive tasks. However, existing trajectory-…
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
Kai Chen, Zichen Ding, Jiaye Ge +20
As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly…
Enhancing LLM-based Search Agents via Contribution Weighted Group Relative Policy Optimization
Junzhe Wang, Zhiheng Xi, Yajie Yang +4
Search agents extend Large Language Models (LLMs) beyond static parametric knowledge by enabling access to up-to-date and long-tail information unavailable during pretraining. Whil…
AgentV-RL: Scaling Reward Modeling with Agentic Verifier
Jiazheng Zhang, Ziche Fu, Zhiheng Xi +13
Verifiers have been demonstrated to enhance LLM reasoning via test-time scaling (TTS). Yet, they face significant challenges in complex domains. Error propagation from incorrect in…
Positive-Only Drifting Policy Optimization
Qi Zhang
In the field of online reinforcement learning (RL), traditional Gaussian policies and flow-based methods are often constrained by their unimodal expressiveness, complex gradient cl…
Step-level Denoising-time Diffusion Alignment with Multiple Objectives
Qi Zhang, Dawei Wang, Shaofeng Zou
Reinforcement learning (RL) has emerged as a powerful tool for aligning diffusion models with human preferences, typically by optimizing a single reward function under a KL regular…
MM-Doc-R1: Training Agents for Long Document Visual Question Answering through Multi-turn Reinforcement Learning
Jiahang Lin, Kai Hu, Binghai Wang +12
Conventional Retrieval-Augmented Generation (RAG) systems often struggle with complex multi-hop queries over long documents due to their single-pass retrieval. We introduce MM-Doc-…
Yet Even Less Is Even Better For Agentic, Reasoning, and Coding LLMs
CodeArts Model Team, Yang Ye, Jingyuan Tan +24
Training effective software engineering agents requires large volumes of task-specific trajectories, incurring substantial data construction costs. Inspired by the "Less-Is-More" h…
Can RL Improve Generalization of LLM Agents? An Empirical Study
Zhiheng Xi, Xin Guo, Jiaqi Liu +11
Reinforcement fine-tuning (RFT) has shown promise for training LLM agents to perform multi-turn decision-making based on environment feedback. However, most existing evaluations re…
JFTA-Bench: Evaluate LLM's Ability of Tracking and Analyzing Malfunctions Using Fault Trees
Yuhui Wang, Zhixiong Yang, Ming Zhang +10
In the maintenance of complex systems, fault trees are used to locate problems and provide targeted solutions. To enable fault trees stored as images to be directly processed by la…
Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment
Yuming Yang, Mingyoung Lai, Wanxu Zhao +13
Long chain-of-thought (CoT) trajectories provide rich supervision signals for distilling reasoning from teacher to student LLMs. However, both prior work and our experiments show t…
TaxoBench: Can Deep Research Agents Retrieve and Organize? Evaluating the Synthesis Gap with Expert Taxonomies
Ming Zhang, Jiabao Zhuang, Wenqing Jing +19
Deep Research Agents increasingly automate survey writing, yet existing benchmarks do not jointly test whether they retrieve the papers experts consider essential and organize thos…
Agri-R1: Agricultural Reasoning for Disease Diagnosis via Automated-Synthesis and Reinforcement Learning
Wentao Zhang, Mingkun Xu, Qi Zhang +6
Agricultural disease diagnosis challenges VLMs, as conventional fine-tuning requires extensive labels, lacks interpretability, and generalizes poorly. While reasoning improves mode…