Supervised fine-tuning (SFT) on a small, high-quality set of long reasoning traces is an effective approach for eliciting strong reasoning capabilities in Large Language Models (LLMs). However, existing methods for curating high-quality SFT data rely heavily on strong reasoning models to filter examples based on diversity and difficulty, making the curation process costly while often yielding suboptimal data quality.
arXiv:2606. 26797v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) on a small, high-quality set of long reasoning traces is an effective approach for eliciting strong reasoning capabilities in Large Language Models (LLMs).
By Hongyi Henry Jin, Wenhan Yang, Meysam Ghaffari, Carlos Morato, Baharan Mirzasoleiman
arXiv:2607. 09999v1 Announce Type: cross Abstract: We show that post-training quantization can silently alter how large language models reason even when task accuracy is preserved.
By Renuka Oladri, Mohan Vamsi Varadaraju Priya, Jerry Wu
The paper introduces CodeRQ-Bench, the first benchmark for assessing large language model reasoning quality across coding tasks such as generation, summarization, and classification. It analyzes over a thousand mismatches from existing evaluators, identifies recurring limitations, and derives design insights that lead to a new two‑stage evaluator, VERA. Experiments show VERA outperforms strong baselines, improving AUCROC by up to 0.26 and AUPRC by up to 0.21 on four datasets.
By Yuangang Li, Justin Tian Jin Chen, Ethan Yu, David Hong, Iftekhar Ahmed
The paper shows that chain‑of‑thought (CoT) instructions can distort multiple‑choice vision‑language model evaluation when a scorer appends a reasoning cue but reads answer‑label logits before the model generates any rationale. This CoT‑prefix scoring causes significant drops in accuracy (e.g., Qwen2.5‑VL‑7B falls from 80.76% to 45.48% on ScienceQA) and leads most predictions to choose the first option. Analysis reveals that while answer information remains linearly accessible in late layers, the immediate readout is misled by probability mass shifting toward continuation tokens, and the issue varies across datasets and models.
The paper shows that chain‑of‑thought (CoT) instructions can distort evaluation of vision‑language models (VLMs) when a scorer reads answer‑label logits before the model generates a rationale. On ScienceQA, Qwen2.5‑VL‑7B’s accuracy falls from 80.76% to 45.48% under this CoT‑prefix scoring, and most predictions incorrectly pick the first option. Linear probes and free generation recover most of the lost accuracy, indicating that the answer information remains in the hidden states but is missed by the early readout. The authors explain the mismatch with vocabulary and layer diagnostics, noting that probability mass shifts toward continuation tokens while answer information stays linearly accessible in later layers. The effect varies across datasets and models, but the study demonstrates that CoT‑prefix scoring can misrepresent model knowledge unless the requested and scored outputs are aligned.
By Zeyan Li, Siyuan Qiu, Jianfeng Xu
arXiv:2604. 08571v3 Announce Type: replace-cross Abstract: While Large Language Models (LLMs) achieve high performance on standard mathematical benchmarks, their problem-solving abilities depend on the context and textual formatting.
By Pavel Golikov, Evgenii Opryshko, Gennady Pekhimenko, Mark C. Jeffrey
The paper introduces OmicsBench, a new reasoning benchmark for multi‑omics sequences that includes 1,160 expert‑validated questions across DNA regulation, RNA processing, and protein function tasks, requiring traceable evidence chains. Evaluation of 17 large language models shows that scientific LLMs, while more accurate in classification, often lack valid evidence, suggesting shortcut learning. To address this, the authors propose tool‑augmented on‑policy distillation (TA‑OPD), a post‑training method that improves both evidence grounding and predictive performance across five Qwen3.5 models of varying sizes.
By Jie Ying, Zhefan Wang, Zihong Chen, Zhengqing Li, Jinzhe Li, Gang Li, Jian Liu, Fang Hu, Tao Luo, Zhonghang Yuan, Wanli Ouyang, Stan Z. Li, Fan Yang, Nanqing Dong
arXiv:2510. 13940v4 Announce Type: replace-cross Abstract: Recent progress in large language models (LLMs) has focused on test-time scaling to improve reasoning via increased inference computation, but often at the cost of efficiency.
By Zhen Yang, Mingyang Zhang, Feng Chen, Ganggui Ding, Liang Hou, Xin Tao, Ying-Cong Chen
arXiv:2607. 21453v1 Announce Type: new Abstract: Scaling inference-time computation has emerged as a reliable method to improve the performance of large language models on complex reasoning and programming tasks.
By Rajiv Shailesh Chitale, Rahul Madhavan, Taneesh Gupta, Deepanway Ghosal, Aravindan Raghuveer
arXiv:2610.29378v1 Announce Type: cross
Abstract: Quantization-aware training (QAT) aims to recover reasoning capabilities in low-bit large language models (LLMs). Existing recipes typically supervis...
By Boyuan Zheng, Jing Xiong, Chenwei Wu, Hui Shen, Zesen Zhao, Hei Ting Una Chan, Xueshen Liu, Zhen Zhang, Liyue Shen, Zechun Liu, Z. Morley Mao
arXiv:2608. 04001v1 Announce Type: cross Abstract: Large language models can solve substantially harder reasoning problems with more inference-time compute.
By Mohsen Hariri, Weicong Chen, Nahal Shahini, Vikash Singh, Kai Ye, Amirhossein Samandar, Debargha Ganguly, Sreehari Sankar, Yanyan Zhang, Shouren Wang, Jerry Peng, Biyao Zhang, Michael Hinczewski, Vipin Chaudhary