English
Related papers

Related papers: Refusal Before Decoding: Detecting and Exploiting …

200 papers

As VLMs are deployed in safety-critical applications, their ability to abstain from answering when uncertain becomes crucial for reliability, especially in Scene Text Visual Question Answering (STVQA) tasks. For example, OCR errors like…

Computer Vision and Pattern Recognition · Computer Science 2025-11-26 Jihan Yao , Achin Kulshrestha , Nathalie Rauschmayr , Reed Roberts , Banghua Zhu , Yulia Tsvetkov , Federico Tombari

Large Language Models (LLMs) are increasingly relied upon for solving complex reasoning tasks in domains such as mathematics, logic, and multi-step question answering. A growing line of work seeks to improve reasoning quality by scaling…

Machine Learning · Computer Science 2025-08-05 Seyyed Saeid Cheshmi , Azal Ahmad Khan , Xinran Wang , Zirui Liu , Ali Anwar

Using responses generated by high-performing large language models (LLMs) for instruction tuning has become a widely adopted approach. However, the existing literature overlooks a property of LLM-generated responses: they conflate world…

Computation and Language · Computer Science 2026-04-16 Tatsuya Ichinose , Youmi Ma , Masanari Oi , Ryuto Koike , Naoaki Okazaki

Large language models (LLMs) increasingly exhibit behaviors suggesting awareness of their evaluation context, often adapting their reasoning strategies in benchmark settings. Prior work has shown that such evaluation awareness can distort…

Computation and Language · Computer Science 2026-05-12 Yanshi Li , Xueru Bai , Shuman Liu , Haibo Zhang , Anxiang Zeng

The field of large language models (LLMs) has grown rapidly in recent years, driven by the desire for better efficiency, interpretability, and safe use. Building on the novel approach of "activation engineering," this study explores…

Computation and Language · Computer Science 2025-08-26 Rumi Allbert , James K. Wiles , Vlad Grankovsky

Prompt-based learning has been demonstrated as a compelling paradigm contributing to large language models' tremendous success (LLMs). Inspired by their success in language tasks, existing research has leveraged LLMs in embodied instruction…

Active learning aims to alleviate the amount of labor involved in data labeling by automating the selection of unlabeled samples via an acquisition function. For example, variational adversarial active learning (VAAL) leverages an…

Machine Learning · Computer Science 2024-08-26 Zongyao Lyu , William J. Beksi

Large language models can detect their own errors and sometimes correct them without external feedback, but the underlying mechanisms remain unknown. We investigate this through the lens of second-order models of confidence from decision…

Machine Learning · Computer Science 2026-05-04 Dharshan Kumaran , Viorica Patraucean , Simon Osindero , Petar Veličković , Nathaniel Daw

While induction is considered a key mechanism for in-context learning in LLMs, understanding its precise circuit decomposition beyond toy models remains elusive. Here, we study the emergence of induction behavior within LLMs by probing…

Machine Learning · Computer Science 2024-11-12 Niclas Luick

Precoding is a key technique for interference management and performance improvement in multi-antenna wireless systems. However, existing precoding methods are typically developed for specific system models, objectives, and constraint sets,…

Multiagent Systems · Computer Science 2026-05-08 Zijiu Yang , Zixiang Zhang , Shunpu Tang , Qianqian Yang , Zhiguo Shi

Large Language Models (LLMs) have demonstrated impressive reasoning capabilities but continue to struggle with arithmetic tasks. Prior works largely focus on outputs or prompting strategies, leaving the open question of the internal…

Computation and Language · Computer Science 2025-11-04 Dharunish Yugeswardeenoo , Harshil Nukala , Ved Shah , Cole Blondin , Sean O Brien , Vasu Sharma , Kevin Zhu

Masked language modeling (MLM) pre-training methods such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. While they produce good results when transferred to…

Computation and Language · Computer Science 2020-03-25 Kevin Clark , Minh-Thang Luong , Quoc V. Le , Christopher D. Manning

We consider the question: when a large language reasoning model makes a choice, did it think first and then decide to, or decide first and then think? In this paper, we present evidence that detectable, early-encoded decisions shape…

Artificial Intelligence · Computer Science 2026-04-06 Esakkivel Esakkiraja , Sai Rajeswar , Denis Akhiyarov , Rajagopal Venkatesaramani

We identify a structural weakness in current large language model (LLM) alignment: modern refusal mechanisms are fail-open. While existing approaches encode refusal behaviors across multiple latent features, suppressing a single dominant…

Machine Learning · Computer Science 2026-02-20 Zachary Coalson , Beth Sohler , Aiden Gabriel , Sanghyun Hong

Multi-turn prompt injection follows a known attack path -- trust-building, pivoting, escalation but text-level defenses miss covert attacks where individual turns appear benign. We show this attack path leaves an activation-level signature…

Cryptography and Security · Computer Science 2026-05-01 Prashant Kulkarni

Despite the outstanding performance of Large language Models (LLMs) in diverse tasks, they are vulnerable to jailbreak attacks, wherein adversarial prompts are crafted to bypass their security mechanisms and elicit unexpected responses.…

Cryptography and Security · Computer Science 2025-04-25 Zeqing He , Zhibo Wang , Zhixuan Chu , Huiyu Xu , Wenhui Zhang , Qinglong Wang , Rui Zheng

Agentic search requires large language models (LLMs) to perform multi-step search to solve complex information-seeking tasks, imposing unique challenges on their reasoning capabilities. However, what constitutes effective reasoning for…

Artificial Intelligence · Computer Science 2026-01-19 Jiahe Jin , Abhijay Paladugu , Chenyan Xiong

Safety alignment in large language models (LLMs), particularly for cybersecurity tasks, primarily focuses on preventing misuse. While this approach reduces direct harm, it obscures a complementary failure mode: denial of assistance to…

Cryptography and Security · Computer Science 2026-03-12 David Campbell , Neil Kale , Udari Madhushani Sehwag , Bert Herring , Nick Price , Dan Borges , Alex Levinson , Christina Q Knight

Binary vulnerability analysis is increasingly performed by LLM-based agents in an iterative, multi-pass manner, with the model as the core decision-maker. However, how such systems organize exploration over hundreds of reasoning steps…

Artificial Intelligence · Computer Science 2026-03-20 Qiang Li , XiangRui Zhang , Haining Wang

Large Language Models (LLMs) are increasingly integrated into code editors to provide AI-powered code suggestions. Yet many of these suggestions are ignored, resulting in wasted computation, increased latency, and unnecessary interruptions.…

Software Engineering · Computer Science 2026-02-10 Mohammad Nour Al Awad , Sergey Ivanov , Olga Tikhonova