XC
Xunliang Cai
cs.CLcs.AIcs.LGcs.CVcs.SEcs.SDcs.HCeess.AScs.DCcs.IR
On Valency
published · living versionsW_kbpfrmbk·v1 · currentpublished
What RL Brings to Reasoning: A Few Decisions the Base Model Would Not Make
with Siyu Zhang, Yuquan Lu, Yoshua Y. Li, Yifan Ding +5
1 version
Preprints & journals
103 papers in the corpus · 2021–2026Beyond Final Scores: A Systematic Evaluation of Agents for Long-Horizon AI Research and Development2608.13417v1 · Yiwei Li, Wanli Yang, Hexiang Tan et al.2026 · 0 citationsarXiv
Contrastive Reinforced Policy Optimization via Privileged Self-Distillation2607.28026v1 · Xingjian Wu, Junlin Liu, Xingchen Liu et al.2026 · 0 citationsarXiv
ClawTrack: Towards Trace-Level Evaluation and Improvement of Real-World Autonomous Agents2607.28037v1 · Xingjian Wu, Xuhang Zhu, Xingchen Liu et al.2026 · 0 citationsarXiv
LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling2606.12837v2 · Jiarui Zhao, Rongzhi Zhang, Lingchuan Liu et al.2026 · 0 citationsarXiv
DailyReport: An Open-ended Benchmark for Evaluating Search Agents on Daily Search Tasks2606.12871v1 · Jingxuan Han, Wei Liu, Mingyang Zhu et al.2026 · 0 citationsarXiv
STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios2606.10394v1 · Sirui Liang, Bohan Yu, Peiyu Wang et al.2026 · 0 citationsarXiv
HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning2606.10507v1 · Juncheng Diao, Zhicong Lu, Peiguang Li et al.2026 · 0 citationsarXiv
LANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint Guidance2605.22567v1 · Yuchun Fan, Bei Li, Peiguang Li et al.2026 · 0 citationsarXiv
MTR-Suite: A Framework for Evaluating and Synthesizing Conversational Retrieval Benchmarks2605.20729v1 · Junhao Ruan, Abudukeyumu Abudula, Bei Li et al.2026 · 0 citationsarXiv
AgentEscapeBench: Evaluating Out-of-Domain Tool-Grounded Reasoning in LLM Agents2605.07926v2 · Zhengkang Guo, Yiyang Li, Lin Qiu et al.2026 · 0 citationsarXiv
Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning2605.06130v3 · Yaorui Shi, Yuxin Chen, Zhengxi Lu et al.2026 · 0 citationsarXiv
Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level2605.06387v4 · Nan Jia, Haojin Yang, Xing Ma et al.2026 · 0 citationsarXiv
AJ-Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation2604.18240v1 · Wentao Shi, Yu Wang, Yuyang Zhao et al.2026 · 0 citationsarXiv
AsyncTLS: Efficient Generative LLM Inference with Asynchronous Two-level Sparse Attention2604.07815v1 · Yuxuan Hu, Jianchao Tan, Jiaqi Zhang et al.2026 · 0 citationsarXiv
Doctor-RAG: A Failure-Aware Repair Framework for Agentic Retrieval-Augmented Generation2604.00865v2 · Shuguang Jiao, Chengkai Huang, Shuhan Qi et al.2026 · 0 citationsarXiv
LongCat-AudioDiT: High-Fidelity Diffusion Text-to-Speech in the Waveform Latent Space2603.29339v1 · Detai Xin, Shujie Hu, Chengzuo Yang et al.2026 · 0 citationsarXiv
LongCat-Next: Lexicalizing Modalities as Discrete Tokens2603.27538v1 · Meituan LongCat Team: Bin Xiao, Chao Wang, Chengjiang Li et al.2026 · 0 citationsarXiv
LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning2603.21065v1 · Jianing Wang, Jianfei Zhang, Qi Guo et al.2026 · 0 citationsarXiv
Can RL Improve Generalization of LLM Agents? An Empirical Study2603.12011v1 · Zhiheng Xi, Xin Guo, Jiaqi Liu et al.2026 · 0 citationsarXiv
Harmonizing Dense and Sparse Signals in Multi-turn RL: Dual-Horizon Credit Assignment for Industrial Sales Agents2603.01481v1 · Haojin Yang, Ai Jian, Xinyue Huang et al.2026 · 0 citationsarXiv
How to Allocate, How to Learn? Dynamic Rollout Allocation and Advantage Modulation for Policy Optimization2602.19208v2 · Yangyi Fang, Jiaye Lin, Xiaoliang Fu et al.2026 · 0 citationsarXiv
MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning2602.17550v3 · Xiaoliang Fu, Jiaye Lin, Yangyi Fang et al.2026 · 0 citationsarXiv
Learning to Self-Verify Makes Language Models Better Reasoners2602.07594v1 · Yuxin Chen, Yu Wang, Yi Zhang et al.2026 · 0 citationsarXiv
DIFFA-2: A Practical Diffusion Large Language Model for General Audio Understanding2601.23161v1 · Jiaming Zhou, Xuxin Cheng, Shiwan Zhao et al.2026 · 0 citationsarXiv
LongCat-Flash-Thinking-2601 Technical Report2601.16725v2 · Meituan LongCat Team, Anchun Gui, Bei Li et al.2026 · 0 citationsarXiv
BAPO: Boundary-Aware Policy Optimization for Reliable Agentic Search2601.11037v2 · Shiyu Liu, Yongjing Yin, Jianhao Yan et al.2026 · 0 citationsarXiv
Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text2601.10355v1 · Zhihao Xu, Rumei Li, Jiahuan Li et al.2026 · 0 citationsarXiv
Efficient Context Scaling with LongCat ZigZag Attention2512.23966v2 · Chen Zhang, Yang Bai, Jiahuan Li et al.2025 · 0 citationsarXiv
EditThinker: Unlocking Iterative Reasoning for Any Image Editor2512.05965v1 · Hongyu Li, Manyuan Zhang, Dian Zheng et al.2025 · 0 citationsarXiv
Scaling and Transferability of Annealing Strategies in Large Language Model Training2512.13705v1 · Siqi Wang, Zhengyu Chen, Teng Xiao et al.2025 · 0 citationsarXiv
A Preliminary Study on the Promises and Challenges of Native Top- Sparse Attention2512.03494v1 · Di Xiu, Hongyin Tang, Bolin Rong et al.2025 · 0 citationsarXiv
OneThinker: All-in-one Reasoning Model for Image and Video2512.03043v3 · Kaituo Feng, Manyuan Zhang, Hongyu Li et al.2025 · 0 citationsarXiv
Friend or Foe: How LLMs' Safety Mind Gets Fooled by Intent Shift Attack2511.00556v1 · Peng Ding, Jun Kuang, Wen Sun et al.2025 · 0 citationsarXiv
AMO-Bench: Large Language Models Still Struggle in High School Math Competitions2510.26768v1 · Shengnan An, Xunliang Cai, Xuezhi Cao et al.2025 · 0 citationsarXiv
CATArena: Evaluating Evolutionary Capabilities of Code Agents via Iterative Tournaments2510.26852v2 · Lingyue Fu, Xin Ding, Linyue Pan et al.2025 · 0 citationsarXiv
PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward Modeling2510.24235v3 · Ai Jian, Jingqing Ruan, Xing Ma et al.2025 · 0 citationsarXiv
Automatically Benchmarking LLM Code Agents through Agent-Driven Annotation and Evaluation2510.24358v3 · Lingyue Fu, Bolun Zhang, Hao Guan et al.2025 · 1 citationProc. of the 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026), Paphos, Cyprus, May 25-29, 2026
Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Views2510.18632v4 · Zhangquan Chen, Manyuan Zhang, Xinlei Yu et al.2025 · 0 citationsCVPR 2026
Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems2510.13291v2 · Xuxin Cheng, Ke Zeng, Zhiquan Cao et al.2025 · 0 citationsarXiv
CodePlot-CoT: Mathematical Visual Reasoning by Thinking with Code-Driven Images2510.11718v1 · Chengqi Duan, Kaiyue Sun, Rongyao Fang et al.2025 · 0 citationsarXiv
Autoformalizer with Tool Feedback2510.06857v1 · Qi Guo, Jianing Wang, Jianfei Zhang et al.2025 · 0 citationsarXiv
AdaR: A Framework for Equipping LLMs with Adaptive Reasoning2510.04617v3 · Zhejian Lai, Xiang Geng, Zhijun Wang et al.2025 · 0 citationsarXiv
MUSE: MCTS-Driven Red Teaming Framework for Enhanced Multi-Turn Dialogue Safety in Large Language Models2509.14651v1 · Siyu Yan, Long Zeng, Xuecheng Wu et al.2025 · 2 citationsarXiv
Instance-level Randomization: Toward More Stable LLM Evaluations2509.12678v1 · Yiyang Li, Yonghuang Wu, Ying Luo et al.2025 · 0 citationsarXiv
WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling2508.16676v2 · Jiacheng Li, Jianchao Tan, Zhidong Yang et al.2025 · 0 citationsarXiv
InfiniteTalk: Audio-driven Video Generation for Sparse-Frame Video Dubbing2508.14033v1 · Shaoshu Yang, Zhe Kong, Feng Gao et al.2025 · 0 citationsarXiv
LinkQA: Synthesizing Diverse QA from Multiple Seeds Strongly Linked by Knowledge Points2508.01317v2 · Xuemiao Zhang, Can Ren, Chengying Tu et al.2025 · 0 citationsarXiv
Large-Scale Diverse Synthesis for Mid-Training2508.01326v1 · Xuemiao Zhang, Chengying Tu, Can Ren et al.2025 · 0 citationsarXiv
CoreCodeBench: Decoupling Code Intelligence via Fine-Grained Repository-Level Tasks2507.05281v2 · Lingyue Fu, Hao Guan, Bolun Zhang et al.2025 · 0 citationsarXiv
Reasoner for Real-World Event Detection: Scaling Reinforcement Learning via Adaptive Perplexity-Aware Sampling Strategy2507.01327v1 · Xiaoyun Zhang, Jingqing Ruan, Xing Ma et al.2025 · 0 citationsarXiv
Career total: 137 works. 103 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.