English
Related papers

Related papers: Language Models Can Predict Their Own Behavior

200 papers

Large language models (LMs) are currently trained to predict tokens given document prefixes, enabling them to directly perform long-form generation and prompting-style tasks which can be reduced to document completion. Existing pretraining…

We investigate whether internal activations in language models can be used to detect arithmetic errors. Starting with a controlled setting of 3-digit addition, we show that simple probes can accurately decode both the model's predicted…

Computation and Language · Computer Science 2025-07-17 Yucheng Sun , Alessandro Stolfo , Mrinmaya Sachan

An important aspect in developing language models that interact with humans is aligning their behavior to be useful and unharmful for their human users. This is usually achieved by tuning the model in a way that enhances desired behaviors…

Computation and Language · Computer Science 2024-06-04 Yotam Wolf , Noam Wies , Oshri Avnery , Yoav Levine , Amnon Shashua

Large language models (LLMs) perform better when they produce step-by-step, "Chain-of-Thought" (CoT) reasoning before answering a question, but it is unclear if the stated reasoning is a faithful explanation of the model's actual reasoning…

We investigate whether the success of a zero-shot Chain-of-Thought (CoT) process can be predicted before completion. We discover that a probing classifier, based on LLM representations, performs well \emph{even before a single token is…

Computation and Language · Computer Science 2025-06-03 Anum Afzal , Florian Matthes , Gal Chechik , Yftah Ziser

Chain-of-Thought (CoT) monitoring has emerged as a compelling method for detecting harmful behaviors such as reward hacking for reasoning models, under the assumption that models' reasoning processes are informative of such behaviors. In…

Machine Learning · Computer Science 2026-03-10 Nikolaus Howe , Micah Carroll

Pre-trained language models (PLMs) may fail in giving reliable estimates of their predictive uncertainty. We take a close look into this problem, aiming to answer two questions: (1) Do PLMs learn to become calibrated in the training…

Computation and Language · Computer Science 2023-05-09 Yangyi Chen , Lifan Yuan , Ganqu Cui , Zhiyuan Liu , Heng Ji

Large Language Models (LLMs) can achieve strong performance on many tasks by producing step-by-step reasoning before giving a final output, often referred to as chain-of-thought reasoning (CoT). It is tempting to interpret these CoT…

Computation and Language · Computer Science 2023-12-12 Miles Turpin , Julian Michael , Ethan Perez , Samuel R. Bowman

Pre-trained language models (LMs) have shown remarkable reasoning performance using explanations or chain-of-thoughts (CoT)) for in-context learning. On the other hand, these reasoning tasks are usually presumed to be more approachable for…

Computation and Language · Computer Science 2024-03-29 Yi-Fan Zhang , Hanlin Zhang , Li Erran Li , Eric Xing

Large language models (LLMs) have grown more powerful in language generation, producing fluent text and even imitating personal style. Yet, this ability also heightens the risk of identity impersonation. To the best of our knowledge, no…

Computation and Language · Computer Science 2026-05-01 Lang Gao , Xuhui Li , Chenxi Wang , Mingzhe Li , Wei Liu , Zirui Song , Jinghui Zhang , Rui Yan , Preslav Nakov , Xiuying Chen

Recent explorations with commercial Large Language Models (LLMs) have shown that non-expert users can jailbreak LLMs by simply manipulating their prompts; resulting in degenerate output behavior, privacy and security breaches, offensive…

Computation and Language · Computer Science 2024-03-28 Abhinav Rao , Sachin Vashistha , Atharva Naik , Somak Aditya , Monojit Choudhury

Language models are susceptible to bias, sycophancy, backdoors, and other tendencies that lead to unfaithful responses to the input context. Interpreting internal states of language models could help monitor and correct unfaithful behavior.…

Computation and Language · Computer Science 2024-12-10 Jiahai Feng , Stuart Russell , Jacob Steinhardt

Autoregressive language models (LMs) generate one token at a time, yet human reasoning operates over higher-level abstractions - sentences, propositions, and concepts. This contrast raises a central question- Can LMs likewise learn to…

Computation and Language · Computer Science 2025-10-14 Hyeonbin Hwang , Byeongguk Jeon , Seungone Kim , Jiyeon Kim , Hoyeon Chang , Sohee Yang , Seungpil Won , Dohaeng Lee , Youbin Ahn , Minjoon Seo

We propose a novel approach to conformal prediction for generative language models (LMs). Standard conformal prediction produces prediction sets -- in place of single predictions -- that have rigorous, statistical performance guarantees. LM…

Computation and Language · Computer Science 2024-06-04 Victor Quach , Adam Fisch , Tal Schuster , Adam Yala , Jae Ho Sohn , Tommi S. Jaakkola , Regina Barzilay

Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe responses, motivating pre-model and post-model guards. Pre-model guards audit the safety of prompts before invoking target models. However,…

Cryptography and Security · Computer Science 2026-05-20 Hongyu Cai , Arjun Arunasalam , Yiming Liang , Antonio Bianchi , Z. Berkay Celik

Large Language Models (LLMs) have started to demonstrate the ability to persuade humans, yet our understanding of how this dynamic transpires is limited. Recent work has used linear probes, lightweight tools for analyzing model…

Computation and Language · Computer Science 2025-08-08 Brandon Jaipersaud , David Krueger , Ekdeep Singh Lubana

Large Language Models (LLMs) have revolutionized artificial intelligence, demonstrating remarkable computational power and linguistic capabilities. However, these models are inherently prone to various biases stemming from their training…

Computation and Language · Computer Science 2025-02-14 Riccardo Cantini , Giada Cosenza , Alessio Orsino , Domenico Talia

Large language models (LLMs) rely on safety alignment to avoid responding to malicious user inputs. Unfortunately, jailbreak can circumvent safety guardrails, resulting in LLMs generating harmful content and raising concerns about LLM…

Computation and Language · Computer Science 2024-06-14 Zhenhong Zhou , Haiyang Yu , Xinghua Zhang , Rongwu Xu , Fei Huang , Yongbin Li

Current research found the issue of Early Answering in large language models (LLMs), where the models already have an answer before generating the Chain-of-Thought (CoT). This phenomenon suggests a potential lack of necessary dependency…

Computation and Language · Computer Science 2025-03-18 Zezhong Wang , Xingshan Zeng , Weiwen Liu , Yufei Wang , Liangyou Li , Yasheng Wang , Lifeng Shang , Xin Jiang , Qun Liu , Kam-Fai Wong

Language models (LMs) are pre-trained on raw text datasets to generate text sequences token-by-token. While this approach facilitates the learning of world knowledge and reasoning, it does not explicitly optimize for linguistic competence.…

Computation and Language · Computer Science 2026-04-17 Atsuki Yamaguchi , Maggie Mi , Nikolaos Aletras