Related papers: Adaptive Instruction Composition for Automated LLM…

Test-time interventions for language models can enhance factual accuracy, mitigate harmful outputs, and improve model efficiency without costly retraining. But despite a flood of new methods, different types of interventions are largely…

This work proposes a procedure for designing algorithms for specific adaptive data collection tasks like active learning and pure-exploration multi-armed bandits. Unlike the design of traditional adaptive algorithms that rely on…

Large visual language models (LVLMs) have demonstrated excellent instruction-following capabilities, yet remain vulnerable to stealthy backdoor attacks when finetuned using contaminated data. Existing backdoor defense techniques are usually…

Advancements in additive manufacturing have enabled design and fabrication of materials and structures not previously realizable. In particular, the design space of composite materials and structures has vastly expanded, and the resulting…

Mix-up training approaches have proven to be effective in improving the generalization ability of Deep Neural Networks. Over the years, the research community expands mix-up methods into two directions, with extensive efforts to improve…

We study how to watermark LLM outputs, i.e. embedding algorithmically detectable signals into LLM-generated text to track misuse. Unlike the current mainstream methods that work with a fixed LLM, we expand the watermark design space by…

Safety alignment in Large Language Models (LLMs) often creates a systematic discrepancy between a model's aligned output and the underlying pre-aligned data distribution. We propose a framework in which the effect of safety alignment on…

Effective training-time guidance is central to multi-agent reinforcement learning (MARL), yet remains difficult in sparse-reward settings where weak supervision limits coordination and policy improvement, and existing methods often require…

Large Language Models (LLMs) often struggle to maintain their original performance when faced with semantically coherent but task-irrelevant contextual information. Although prior studies have explored this issue using fixed-template or…

Large language models (LLMs) are vulnerable when trained on datasets containing harmful content, which leads to potential jailbreaking attacks in two scenarios: the integration of harmful texts within crowdsourced data used for pre-training…

Embeddings extracted by pre-trained Large Language Models (LLMs) have significant potential to improve information retrieval and search. Beyond the zero-shot setup in which they are being conventionally used, being able to take advantage of…

Large Language Models (LLMs) are trained to refuse harmful requests, yet they remain vulnerable to jailbreak attacks that exploit weaknesses in conversational safety mechanisms. We introduce Incremental Completion Decomposition (ICD), a…

While pre-trained multimodal representations (e.g., CLIP) have shown impressive capabilities, they exhibit significant compositional vulnerabilities leading to counterintuitive judgments. We introduce Multimodal Adversarial Compositionality…

With the advent of the information explosion era, the importance of recommendation systems in various applications is increasingly significant. Traditional collaborative filtering algorithms are widely used due to their effectiveness in…

Agentic language model (LM) systems power modern applications like "Deep Research" and "Claude Code," and leverage multi-LM architectures to overcome context limitations. Beneath their apparent diversity lies a recurring pattern: smaller…

Despite the surprising few-shot performance of in-context learning (ICL), it is still a common practice to randomly sample examples to serve as context. This paper advocates a new principle for ICL: self-adaptive in-context learning. The…

A hallmark of human intelligence is the ability to construct self-contained chunks of knowledge and reuse them in novel combinations for solving different problems. Learning such compositional structures has been a challenge for artificial…

With the advent of Large Language Models (LLMs), generating rule-based data for real-world applications has become more accessible. Due to the inherent ambiguity of natural language and the complexity of rule sets, especially in long…

Jailbreaking attacks on large language models pose a significant threat to AI safety by enabling the generation of harmful or restricted content. While prior work has explored both handcrafted and automated jailbreak strategies, the…

Despite achieving remarkable success in complex tasks, Deep Reinforcement Learning (DRL) is still suffering from critical issues in practical applications, such as low data efficiency, lack of interpretability, and limited cross-environment…