paper-with-me

홈 › Papers

Better Accuracies, Worse Reasoning: A Step-Level Audit of Medical Chain-of-Thought Distillation

2026-05-27 · Zhaoyang Jiang, Xuanqi Peng, Fei Teng, Zhizhong Fu, Yunsoo Kim, Jiacong Mi, Zicheng Li, Honghan Wu arxiv

Chain-of-thought (CoT) distillation trains a smaller model to imitate a teacher's reasoning trace, but it is typically evaluated by final-answer metrics including accuracy. We ask whether gains in answer quality are accompanied by improvements in the trace. In medical QA, where short answer options can leave a richer clinical justification under-specified, a Qwen3-8B student distilled from a DeepSeek-V3-family teacher improves on MedQA-USMLE answer metrics (SC@64 74.7% to 84.4%; expected calibration error (ECE) 0.096 to 0.034). Yet under a Kimi-K2.6 style-blind LLM-judge audit, its error rate over non-abstained steps rises from 30.6% to 50.3%. In this primary medical setting, answer quality and trace factuality move in opposite directions. This before--after pattern persists across evaluators, teacher strengths, student scales and families, medical benchmarks, and style, segmentation, and answer-correctness controls. A 150-step blinded audit by a clinical expert reproduces the same ordering. Boundary checks narrow the scope of the claim: the risk appears when a compact answer under-constrains the rationale and a capable student can imitate expert-like form without reliably grounding each local claim. Standard answer metrics and aggregate hedging rates do not reveal the shift. When such traces are released or reused, answer-level metrics alone are insufficient.

📄 PDF Abstract BibTeX arXiv:2605.28301

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Character-Level Language Modeling with Hierarchical Recurrent Neural Networks

2016-09-13 · Kyuyeon Hwang, Wonyong Sung

Recurrent neural network (RNN) based character-level language models (CLMs) are extremely useful for modeling out-of-vocabulary words by nature. However, their performance is generally much worse than the word-level lang…

Language ModelingLanguage Modellingspeech-recognitionSpeech Recognition

Do Large Language Models Know What They Are Capable Of?

2025-12-31 · Casey O. Barkan, Sid Black, Oliver Sourbut arxiv

We investigate whether large language models (LLMs) can predict whether they will succeed on a given task and whether their predictions improve as they progress through multi-step tasks. We also investigate whether LLMs …

Decision Making

Can AI Master Construction Management (CM)? Benchmarking State-of-the-Art Large Language Models on CM Certification Exams

2025-04-04 · Ruoxin Xiong, Yanyu Wang, Suat Gunhan, Yimin Zhu 외

The growing complexity of construction management (CM) projects, coupled with challenges such as strict regulatory requirements and labor shortages, requires specialized analytical tools that streamline project workflow …

BenchmarkingManagementMultiple-choice

Do better ImageNet classifiers assess perceptual similarity better?

2022-03-09 · Manoj Kumar, Neil Houlsby, Nal Kalchbrenner, Ekin D. Cubuk

Perceptual distances between images, as measured in the space of pre-trained deep features, have outperformed prior low-level, pixel-based metrics on assessing perceptual similarity. While the capabilities of older and l…

Where on Earth? A Vision-Language Benchmark for Probing Model Geolocation Skills Across Scales

2025-10-13 · Zhaofang Qian, Hardy Chen, Zeyu Wang, Li Zhang 외 arxiv

Vision-language models (VLMs) have advanced rapidly, yet their capacity for image-grounded geolocation in open-world conditions, a task that is challenging and of demand in real life, has not been comprehensively evaluat…