相关论文: Language-Switching Triggers Take a Latent Detour T…

We propose a new architectural change, and post-training pipeline, for making LLMs more verbose reasoners by teaching a model to truncate forward passes early. We augment an existing transformer architecture with an early-exit mechanism at…

The real-world information sources are inherently multilingual, which naturally raises a question about whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop question answering…

Metacognition -- assessing the quality of one's own cognitive performance -- guides adaptive behavior across species. Substantial research demonstrates that confidence signals can be extracted from language model outputs, yet a fundamental…

English pretrained language models, which make up the backbone of many modern NLP systems, require huge amounts of unlabeled training data. These models are generally presented as being trained only on English text but have been found to…

Backdoor attacks, in which a model behaves maliciously when given an attacker-specified trigger, pose a major security risk for practitioners who depend on publicly released language models. As a countermeasure, backdoor detection methods…

Recently, ChatGPT has gained significant attention in research due to its ability to interact with humans effectively. The core idea behind this model is reinforcement learning (RL) fine-tuning, a new paradigm that allows language models to…

Explainable machine learning holds great potential for analyzing and understanding learning-based systems. These methods can, however, be manipulated to present unfaithful explanations, giving rise to powerful and stealthy adversaries. In…

Large vision-language models (LVLMs) have achieved impressive performance across a wide range of vision-language tasks, while they remain vulnerable to backdoor attacks. Existing backdoor attacks on LVLMs aim to force the victim model to…

Multilingual language models (MLMs) store factual knowledge across languages but often struggle to provide consistent responses to semantically equivalent prompts in different languages. While previous studies point out this cross-lingual…

Large language models often reason beyond surface tokens, but the internal stage at which token-level information becomes abstract relational structure remains unclear. We investigate this question by analyzing how attention heads and…

Large Language Models (LLMs) are increasingly deployed across multilingual applications that handle sensitive data, yet their scale and linguistic variability introduce major privacy risks. Mostly evaluated for English, this paper…

Due to the high cost of training, large model (LM) practitioners commonly use pretrained models downloaded from untrusted sources, which could lead to owning compromised models. In-context learning is the ability of LMs to perform multiple…

With the rapid rise of personalized AI, customized large language models (LLMs) equipped with Chain of Thought (COT) reasoning now power millions of AI agents. However, their complex reasoning processes introduce new and largely unexplored…

Reasoning has become a central capability in large language models. Recent research has shown that reasoning performance can be improved by looping an LLM's layers in the latent dimension, resulting in looped reasoning language models.…

Backdoor Attacks have been a serious vulnerability against Large Language Models (LLMs). However, previous methods only reveal such risk in specific models, or present tasks transferability after attacking the pre-trained phase. So, how…

Despite significant advances in alignment techniques, we demonstrate that state-of-the-art language models remain vulnerable to carefully crafted conversational scenarios that can induce various forms of misalignment without explicit…

Looped Transformers offer a promising alternative to purely feed-forward computation by iteratively refining latent representations, improving language modeling and reasoning. Yet recurrent architectures remain unstable to train, costly to…

Mainstream backdoor attacks on large language models (LLMs) typically set a fixed trigger in the input instance and specific responses for triggered queries. However, the fixed trigger setting (e.g., unusual words) may be easily detected by…

Humans are capable of strategically deceptive behavior: behaving helpfully in most situations, but then behaving very differently in order to pursue alternative objectives when given the opportunity. If an AI system learned such a deceptive…

Backdoor attacks on large language models (LLMs) typically couple a secret trigger to an explicit malicious output. We show that this explicit association is unnecessary for common LLMs. We introduce a compliance-only backdoor: supervised…