18 papers
SERO: Learning from Executable Oracle Continuations for Long-Horizon Agentic Reinforcement Learning
Dingyi Rong, Yue Shi, Chaofan Ma +5
Long-horizon agentic RL is constrained by the futures visited by the current policy. When rollouts fail to reach useful behaviors, improved credit assignment cannot provide supervi…
Environments as Scaffold: Enriching Feedback to Bootstrap Self-Evolving Agents in Long-Horizon Tasks
Hongbang Yuan, Zhuoran Jin, Yixin Cao
Large Language Models demonstrate remarkable proficiency in static reasoning, yet training them as autonomous agents through Reinforcement Learning (RL) for long-horizon tasks is o…
From Base Rollouts to RL Reasoning: A Budgeted Search Perspective
Wenhe Sun, Cunxiang Wang, Zijun Yao +1
Reinforcement learning with verifiable rewards (RLVR) improves language-model reasoning, but how these gains relate to inference-time decoding and search remains unclear. Does RL c…
ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders
Zhongyuan Peng, Dan Huang, Chuyu Zhang +8
The recent emergence of vibe-coding workflows is changing what coding agents are expected to do. Instead of merely completing code under fully specified instructions, agents are in…
Does the Same Token Mean the Same State? MoE Routing as Signal for Reasoning Control
Kang Chen, Minshen Yu, Junjie Nian +3
In sparse Mixture-of-Experts language models, does the same token id imply the same router state and the same experts producing it? Holding the emitted token id fixed at repeated a…
DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy Prefixes
Caijun Xu, Changyi Xiao, Zhongyuan Peng +1
Reinforcement learning has become a central paradigm for advancing reasoning in large language models, yet most existing methods still depend on stronger teacher models or heavily…
Reinforcement Learning with Conditional Expectation Reward
Changyi Xiao, Caijun Xu, Yixin Cao
Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective in enhancing the reasoning capabilities of large language models, particularly in domains such as mathema…
GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation
Yifan Wang, Mingxuan Jiang, Zhihao Sun +5
Retrieval-Augmented Generation (RAG) grounds large language models with external evidence, but many implementations rely on pre-built indices that remain static after construction.…
CoDiQ: Test-Time Scaling for Controllable Difficult Question Generation
Zhongyuan Peng, Caijun Xu, Changyi Xiao +4
Large Reasoning Models (LRMs) benefit substantially from training on challenging competition-level questions. However, existing automated question synthesis methods lack precise di…
DLEBench: Evaluating Small-scale Object Editing Ability for Instruction-based Image Editing Model
Shibo Hong, Boxian Ai, Jun Kuang +5
Significant progress has been made in the field of Instruction-based Image Editing Models (IIEMs). However, while these models demonstrate plausible adherence to instructions and s…
Retrieval-Infused Reasoning Sandbox: A Benchmark for Decoupling Retrieval and Reasoning Capabilities
Shuangshuang Ying, Zheyu Wang, Yunjian Peng +16
Despite strong performance on existing benchmarks, it remains unclear whether large language models can reason over genuinely novel scientific information. Most evaluations score e…
ARM: Role-Conditioned Neuron Transplantation for Training-Free Generalist LLM Agent Merging
Zhuoka Feng, Kang Chen, Sihan Zhao +7
Interactive large language model agents have advanced rapidly, but most remain specialized to a single environment and fail to adapt robustly to other environments. Model merging o…
SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning
Caijun Xu, Changyi Xiao, Zhongyuan Peng +2
Reinforcement learning (RL) offers a principled way to enhance the reasoning capabilities of large language models, yet its effectiveness hinges on training signals that remain inf…
CriticLean: Critic-Guided Reinforcement Learning for Mathematical Formalization
Zhongyuan Peng, Yifan Yao, Kaijing Ma +16
Translating natural language mathematical statements into formal, executable code is a fundamental challenge in automated theorem proving. While prior work has focused on generatio…
Long or short CoT? Investigating Instance-level Switch of Large Reasoning Models
Ruiqi Zhang, Changyi Xiao, Yixin Cao
With the rapid advancement of large reasoning models, long Chain-of-Thought (CoT) prompting has demonstrated strong performance on complex tasks. However, this often comes with a s…
BNPO: Beta Normalization Policy Optimization
Changyi Xiao, Mengdi Zhang, Yixin Cao
Recent studies, including DeepSeek-R1 and Kimi-k1.5, have demonstrated that reinforcement learning with rule-based, binary-valued reward functions can significantly enhance the rea…
Knowledge Graph Completion by Intermediate Variables Regularization
Changyi Xiao, Yixin Cao
Knowledge graph completion (KGC) can be framed as a 3-order binary tensor completion task. Tensor decomposition-based (TDB) models have demonstrated strong performance in KGC. In t…
Complex Logical Query Answering by Calibrating Knowledge Graph Completion Models
Changyi Xiao, Yixin Cao
Complex logical query answering (CLQA) is a challenging task that involves finding answer entities for complex logical queries over incomplete knowledge graphs (KGs). Previous rese…