Hugging Face Trending Papers August 6, 2026

Learning When to Trust via Selective Context Preference Optimization

Read the original on Hugging Face Trending Papers →

Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong. The obvious remedy, training models to resist such signals, hides a failure mode: a model that ignores all context looks robust yet is useless when the context is worth trusting.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv AI

Aug 7

Learning When to Trust via Selective Context Preference Optimization

arXiv:2608. 06377v1 Announce Type: cross Abstract: Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong.

By Xian Sun, Wei Chow, Yingshuo Wang, Junhao Liu, Wei Gao, Qing Wu, Lingdong Kong

llms reinforcement-learning benchmarks

arXiv AI

Jun 24

Breaking the Mirror: Activation-Based Mitigation of Self-Preference in LLM Evaluators

arXiv:2509. 03647v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly serve as automated evaluators, yet they suffer from "self-preference bias": a tendency to favor their own outputs over those of other models.

By Dani Roytburg, Matthew Bozoukov, Matthew Nguyen, Jou Barzdukas, Simon Fu, Narmeen Oozeer

llms safety

arXiv AI

Jun 2

Consistency Training while Mitigating Obfuscation via Rate Matching

arXiv:2606. 02211v1 Announce Type: cross Abstract: Large language models are often influenced by extraneous input features, such as cues revealing a user's preferred answer.

By Sohaib Imran, Prakhar Gupta, Jannes Elstner, David Demitri Africa

llms safety

arXiv AI

Jul 23

Information Discernment in Large Language Models

arXiv:2607. 19355v1 Announce Type: new Abstract: LLMs are increasingly used with external knowledge sources like the internet.

By Joshua Ashkinaze, Laura Kurek, Alina Faisal, Tongyuan Miao, Mariam Joseph, Ceren Budak, Eric Gilbert

llms benchmarks safety

arXiv AI

Jun 2

One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models

arXiv:2603. 03291v2 Announce Type: replace-cross Abstract: Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences.

By Daniel Fein, Max Lamparth, Violet Xiang, Mykel J. Kochenderfer, Nick Haber

llms benchmarks safety

arXiv AI

1d ago

When Context Misleads: Intent-Guided Decoding for Robust Retrieval-Augmented Generation

arXiv:2608. 16515v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) improves large language models by grounding generation in external evidence, but it also introduces a source trust problem: retrieved context may be useful, irrelevant, or even misleading.

By Haolin Jin, Pengyue Yang, Huaming Chen

llms rag benchmarks

Related stories

Learning When to Trust via Selective Context Preference Optimization

Breaking the Mirror: Activation-Based Mitigation of Self-Preference in LLM Evaluators

Consistency Training while Mitigating Obfuscation via Rate Matching

Information Discernment in Large Language Models

One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models

When Context Misleads: Intent-Guided Decoding for Robust Retrieval-Augmented Generation