作者位次 #4/4ICLR 2026Poster
DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference
Xiang Liu · Xuming Hu · Xiaowen Chu · Eunsol Choi
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Xiang Liu · Xuming Hu · Xiaowen Chu · Eunsol Choi
Nishant Balepur · Bhavya Rajasekaran · Hyunjin Jane Oh · Michael Xie · Atrey Desai · Vipul Gupta · Steven James Moore · Eunsol Choi · Rachel Rudinger · Jordan Lee Boyd-Graber
Multiple-choice question answering (MCQA) is standard in NLP, but benchmarks lack rigorous quality control. We present BenchMarker, an education-inspired toolkit using LLM judges to flag three common MCQ flaws: 1) contamination—items appearing exactly online; 2) shortcuts—cues in the choices that...
Nishant Balepur · Malachi Hamada · Varsha Kishore · Sergey Feldman · Amanpreet Singh · Pao Siangliulue · Joseph Chee Chang · Eunsol Choi · Jordan Lee Boyd-Graber · Aakanksha Naik
Deep Research (DR) tools (e.g. OpenAI DR) help researchers cope with ballooning publishing counts. Such tools can synthesize scientific papers to answer researchers’ queries, but lack understanding of their users. We change that in MyScholarQA (MySQA), a personalized DR tool that: 1) infers a pro...