7 papers
SafeSteer: Localized On-Policy Distillation for Efficient Safety Alignment
Hao Li, Jingkun An, Zijun Song +8
Aligning Large Language Models (LLMs) with human values often degrades their general capabilities, termed the alignment tax. Existing methods mitigate this by balancing dual object…
Auto-Rubric as Reward: From Implicit Preferences to Explicit Multimodal Generative Criteria
Juanxi Tian, Fengyuan Liu, Jiaming Han +6
Aligning multimodal generative models with human preferences demands reward signals that respect the compositional, multi-dimensional structure of human judgment. Prevailing RLHF a…
AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery
Lei Xiong, Kun Luo, Ziyi Xia +15
Autonomous scientific research is significantly advanced thanks to the development of AI agents. One key step in this process is finding the right scientific literature, whether to…
Do Vision-Language Models Measure Up? Benchmarking Visual Measurement Reading with MeasureBench
Fenfen Lin, Yesheng Liu, Haiyu Xu +7
Reading measurement instruments is effortless for humans and requires relatively little domain expertise, yet it remains surprisingly challenging for current vision-language models…
FlagEval Findings Report: A Preliminary Evaluation of Large Reasoning Models on Automatically Verifiable Textual and Visual Questions
Bowen Qin, Chen Yue, Fang Yin +26
We conduct a moderate-scale contamination-free (to some extent) evaluation of current large reasoning models (LRMs) with some preliminary findings. We also release ROME, our evalua…
Beyond Multiple Choice: Verifiable OpenQA for Robust Vision-Language RFT
Yesheng Liu, Hao Li, Haiyu Xu +9
Multiple-choice question answering (MCQA) has been a popular format for evaluating and reinforcement fine-tuning (RFT) of modern multimodal language models. Its constrained output…