相关论文: Entropy-Based Measurement of Value Drift and Align…

Preserving biodiversity and ecosystem stability is a challenge that can be pursued through modern statistical mechanics modeling. Here we introduce a variational maximum entropy-based algorithm to evaluate the entropy in a minimal ecosystem…

Large language models frequently exhibit suboptimal performance on low resource languages, primarily due to inefficient subword segmentation and systemic training data imbalances. In this paper, we propose Variable Entropy Policy…

Multi-step processes via large language models (LLMs) have proven effective for solving complex reasoning tasks. However, the depth of exploration of the reasoning procedure can significantly affect the task performance. Existing methods to…

Reinforcement Learning with Verifiable Rewards (RLVR) for Multimodal Large Language Models (MLLMs) is highly dependent on high-quality labeled data, which is often scarce and prone to substantial annotation noise in real-world scenarios.…

Large language models (LLMs) have been widely used for mental health support. However, current safety evaluations in this field are mostly limited to detecting whether LLMs output prohibited words in single-turn conversations, neglecting…

Understanding how large language models (LLMs) process emotionally sensitive content is critical for building safe and reliable systems, particularly in mental health contexts. We investigate the scaling behavior of LLMs on two key tasks:…

Large Language Models (LLMs) have shown promise as intelligent agents in interactive decision-making tasks. Traditional approaches often depend on meticulously designed prompts, high-quality examples, or additional reward models for…

Reinforcement learning (RL) plays a crucial role in shaping the behavior of large language and reasoning models (LLMs/LRMs). However, it often produces brittle and unstable policies, leading to critical failures such as spurious reasoning,…

Humans display significant uncertainty when confronted with moral dilemmas, yet the extent of such uncertainty in machines and AI agents remains underexplored. Recent studies have confirmed the overly confident tendencies of…

Conversational Large Language Models are post-trained on language that expresses specific behavioural traits, such as curiosity, open-mindedness, and empathy, and values, such as helpfulness, harmlessness, and honesty. This is done to…

As Large Language Models increasingly mediate human communication and decision-making, understanding their value expression becomes critical for research across disciplines. This work presents the Ethics Engine, a modular Python pipeline…

Safety alignment is crucial to ensure that large language models (LLMs) behave in ways that align with human preferences and prevent harmful actions during inference. However, recent studies show that the alignment can be easily compromised…

Large Language Models (LLMs) can be backdoored to exhibit malicious behavior under specific deployment conditions while appearing safe during training a phenomenon known as "sleeper agents." Recent work by Hubinger et al. demonstrated that…

Safety alignment of Large Language Models (LLMs) is extremely fragile, as fine-tuning on a small number of benign samples can erase safety behaviors learned from millions of preference examples. Existing studies attempt to explain this…

Recent advances in large language models (LLMs) have demonstrated significant potential in hardware design automation, particularly in using natural language to synthesize Register-Transfer Level (RTL) code. Despite this progress, a gap…

We propose ERA, a new paradigm that constrains the sampling entropy above given thresholds by applying specially designed activations to the outputs of models. Our approach demonstrates broad effectiveness across different domains: 1) for…

Most adversarial evaluations of large language model (LLM) safety assess single prompts and report binary pass/fail outcomes, which fails to capture how safety properties evolve under sustained adversarial interaction. We present ADVERSA,…

Large language models (LLMs) are increasingly used in social science simulations. While their performance on reasoning and optimization tasks has been extensively evaluated, less attention has been paid to their ability to simulate human…

Large Language Models (LLMs) that can express interpretable and calibrated uncertainty are crucial in high-stakes domains. While methods to compute uncertainty post-hoc exist, they are often sampling-based and therefore computationally…

Large Language Models (LLMs) excel on many task-specific benchmarks, yet the mechanisms that drive this success remain poorly understood. We move from asking what these systems can do to asking how they process information. Our contribution…