9 papers
FinFIRST: Benchmarking Search Agents for Financial Information Retrieval, Sourcing and Traceability
Wenqing Wang, Haitao Xiang, Xinyi Zhao +8
Financial search is a highly demanding task for LLM agents, requiring not only a correct final answer but also temporally valid information retrieval, authoritative source selectio…
Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling
Xinmu Ge, Zizhuo Zhang, Yu Huang +9
On-policy distillation (OPD) has emerged as a promising post-training technique for enhancing LLM reasoning. Under the reverse KL objective, the idealized optimum of OPD aligns the…
FinReportBench: Measuring and Improving Institution-Grade Financial Report Generation
Yinghao Tang, Tan Zhenwei, Yiyao Wang +4
Large language models can produce fluent financial analysis, but fluency alone does not establish whether a report is suitable for institutional delivery. We introduce FinReportBen…
LiveEvalBench: Toward Open-World Evaluation for Web Generation
Yiyao Wang, Zhen Wen, Yinghao Tang +5
Large language models are increasingly capable of synthesizing executable frontend projects, yet existing benchmarks still treat web generation as a static evaluation problem. We a…
Focal Reward: Balanced Reinforcement Learning under Rubric-Based Rewards
Yu Huang, Zihua Zhao, Zhaoxin Huan +9
The open-ended generation in LLMs usually requires multi-dimensional rubrics to adequately assess quality and guide the improvement of reinforcement learning. However, a critical d…
LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion
Guanghao Zhou, Panjia Qiu, Cen Chen +4
The safety mechanisms of large language models (LLMs) exhibit notable fragility, as even fine-tuning on datasets without harmful content may still undermine their safety capabiliti…
Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary
Licheng Pan, Yongqi Tong, Xin Zhang +3
Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks, yet they often refuse to answer legitimate queries--a phenomenon known as overr…
REWARD CONSISTENCY: Improving Multi-Objective Alignment from a Data-Centric Perspective
Zhihao Xu, Yongqi Tong, Xin Zhang +2
Multi-objective preference alignment in language models often encounters a challenging trade-off: optimizing for one human preference (e.g., helpfulness) frequently compromises oth…
CCJA: Context-Coherent Jailbreak Attack for Aligned Large Language Models
Guanghao Zhou, Panjia Qiu, Mingyuan Fan +4
Despite explicit alignment efforts for large language models (LLMs), they can still be exploited to trigger unintended behaviors, a phenomenon known as "jailbreaking." Current jail…