paper-with-me

Papers

Inferring from Logits: Exploring Best Practices for Decoding-Free Generative Candidate Selection

2025-01-28 · Mingyu Derek Ma, Yanna Ding, Zijie Huang, Jianxi Gao, Yizhou Sun, Wei Wang

Generative Language Models rely on autoregressive decoding to produce the output sequence token by token. Many tasks such as preference optimization, require the model to produce task-level output consisting of multiple tokens directly by selecting candidates from a pool as predictions. Determining a task-level prediction from candidates using the ordinary token-level decoding mechanism is constrained by time-consuming decoding and interrupted gradients by discrete token selection. Existing works have been using decoding-free candidate selection methods to obtain candidate probability from initial output logits over vocabulary. Though these estimation methods are widely used, they are not systematically evaluated, especially on end tasks. We introduce an evaluation of a comprehensive collection of decoding-free candidate selection approaches on a comprehensive set of tasks, including five multiple-choice QA tasks with a small candidate pool and four clinical decision tasks with a massive amount of candidates, some with 10k+ options. We evaluate the estimation methods paired with a wide spectrum of foundation LMs covering different architectures, sizes and training paradigms. The results and insights from our analysis inform the future model design.

📄 PDF Abstract BibTeX arXiv:2501.17338

Code (0)

등록된 구현이 없습니다.

Tasks

Multiple-choice

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code Generation

2026-02-15 · Jizheng Chen, Weiming Zhang, Xinyi Dai, Weiwen Liu 외 arxiv

Code generation remains a challenging task that requires precise and structured reasoning. Existing Test Time Scaling (TTS) methods, including structured tree search, have made progress in exploring reasoning paths but s…

Code Generation

CHAIR -- Classifier of Hallucination as Improver

2025-01-05 · Ao Sun

In this work, we introduce CHAIR (Classifier of Hallucination As ImproveR), a supervised framework for detecting hallucinations by analyzing internal logits from each layer of every token. Our method extracts a compact s…

HallucinationMMLUTruthfulQA

Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation

2024-12-05 · Hao Yang, Qianghua Zhao, Lei LI

Chain-of-Thought prompting has significantly enhanced the reasoning capabilities of large language models, with numerous studies exploring factors influencing its performance. However, the underlying mechanisms remain po…

Retrieval

Mitigating Hallucination of Large Vision-Language Models via Dynamic Logits Calibration

2025-06-26 · Jiahe Chen, Jiaying He, Qian Shao, Qiyuan Chen 외

Large Vision-Language Models (LVLMs) have demonstrated significant advancements in multimodal understanding, yet they are frequently hampered by hallucination-the generation of text that contradicts visual input. Existin…

HallucinationText Generation

PruneCD: Contrasting Pruned Self Model to Improve Decoding Factuality

2025-09-20 · Byeongho Yu, Changhun Lee, Jungyu Jin, Eunhyeok Park arxiv

To mitigate the hallucination problem in large language models, DoLa exploits early exit logits from the same model as a contrastive prior. However, we found that these early exit logits tend to be flat, low in magnitude…