10 papers
SAKI: Maximal-Coupling-Routed Teacher Supervision for On-Policy Distillation
Miteto Wei, Xiaohan Wang, Zehao Chen +7
On-policy distillation (OPD) reduces train-test state mismatch by training a student on its own generated trajectories, but weak students may visit teacher-misaligned prefixes wher…
Toward Plasticity-Preserving KL Regularization for Capability Retention in LLM Reinforcement Learning
Li Wang, Xiaodong Lu, Xiaohan Wang +4
Reinforcement learning (RL) has become a central paradigm for large language model (LLM) post-training, but optimization toward new objectives can degrade capabilities already pres…
HybridCodeAuthorship: A Benchmark Dataset for Line-Level Code Authorship Detection
Luke Patterson, Li Wang, Adam Faulkner
Thanks to the rapid adoption of AI code assistants powered by large language models (LLMs), industry codebases are, increasingly, a hybrid of AI- and human-authored code. For risk…
ZipRL: Adaptive Multi-Turn Context Compression with Hindsight Response Replay
Zhexin Hu, Li Wang, Xiaohan Wang +4
Adaptive context compression is vital for scaling Large Language Models (LLMs) to complex, multi-turn agent tasks. However, rule-based compression methods may discard task-critical…
Implicit Hierarchical GRPO: Decoupling Tool Invocation from Execution for Tool-Integrated Mathematical Reasoning
Li Wang, Xiaohan Wang, Xiaodong Lu +5
Large language models (LLMs) have increasingly leveraged tool invocation to enhance their reasoning capabilities. However, existing approaches typically tightly couple tool invocat…
ResRL: Boosting LLM Reasoning via Negative Sample Projection Residual Reinforcement Learning
Zihan Lin, Xiaohan Wang, Jie Cao +6
Reinforcement Learning with Verifiable Rewards (RLVR) enhances reasoning of Large Language Models (LLMs) but usually exhibits limited generation diversity due to the over-incentivi…
Deconstructing Instruction-Following: A New Benchmark for Granular Evaluation of Large Language Model Instruction Compliance Abilities
Alberto Purpura, Li Wang, Sahil Badyal +2
Reliably ensuring Large Language Models (LLMs) follow complex instructions is a critical challenge, as existing benchmarks often fail to reflect real-world use or isolate complianc…
Enhancing LLM Instruction Following: An Evaluation-Driven Multi-Agentic Workflow for Prompt Instructions Optimization
Alberto Purpura, Li Wang, Sahil Badyal +2
Large Language Models (LLMs) often generate substantively relevant content but fail to adhere to formal constraints, leading to outputs that are conceptually correct but procedural…
NCV: A Node-Wise Consistency Verification Approach for Low-Cost Structured Error Localization in LLM Reasoning
Yulong Zhang, Li Wang, Wei Du +7
Verifying multi-step reasoning in large language models is difficult due to imprecise error localization and high token costs. Existing methods either assess entire reasoning chain…
Veracity: An Open-Source AI Fact-Checking System
Taylor Lynn Curtis, Maximilian Puelma Touzel, William Garneau +8
The proliferation of misinformation poses a significant threat to society, exacerbated by the capabilities of generative AI. This demo paper introduces Veracity, an open-source AI…