Related papers: Decomposed Prompting Does Not Fix Knowledge Gaps, …
Despite demonstrating impressive capabilities, Large Language Models (LLMs) still often struggle to accurately express the factual knowledge they possess, especially in cases where the LLMs' knowledge boundaries are ambiguous. To improve…
Large language models (LLMs) are increasingly used for long-document question answering, where reliable attribution to sources is critical for trust. Existing post-hoc attribution methods work well for extractive QA but struggle in…
The text generated by large language models is commonly controlled by prompting, where a prompt prepended to a user's query guides the model's output. The prompts used by companies to guide their models are often treated as secrets, to be…
Deployed language models must decide not only what to answer but also when not to answer. We present UniCR, a unified framework that turns heterogeneous uncertainty evidence including sequence likelihoods, self-consistency dispersion,…
We study the feasibility of identifying epistemic uncertainty (reflecting a lack of knowledge), as opposed to aleatoric uncertainty (reflecting entropy in the underlying distribution), in the outputs of large language models (LLMs) over…
Most large language models (LLMs) are trained once and never updated; thus, they lack the ability to dynamically adapt to our ever-changing world. In this work, we perform a detailed study of the factuality of LLM-generated text in the…
Large Language Models (LLMs) are powerful linguistic engines but remain susceptible to hallucinations: plausible-sounding outputs that are factually incorrect or unsupported. In this work, we present a mathematically grounded framework to…
Large language models excel at complex instructions yet struggle to deviate from their helpful assistant persona, as post-training instills strong priors that resist conflicting instructions. We introduce system prompt strength, a…
Many recent prompting strategies for large language models (LLMs) query the model multiple times sequentially -- first to produce intermediate results and then the final answer. However, using these methods, both decoder and model are…
Large language models are successful in answering factoid questions but are also prone to hallucination. We investigate the phenomenon of LLMs possessing correct answer knowledge yet still hallucinating from the perspective of inference…
While large language models (LLMs) excel in mathematical and code reasoning, we observe they struggle with social reasoning tasks, exhibiting cognitive confusion, logical inconsistencies, and conflation between objective world states and…
Hallucinations in large language models (LLMs) present a growing challenge across real-world applications, from healthcare to law, where factual reliability is essential. Despite advances in alignment and instruction tuning, LLMs can still…
Fact-seeking question answering with large language models (LLMs) remains unreliable when answers depend on up-to-date or conflicting information. Although retrieval-augmented and tool-using LLMs reduce hallucinations, they often rely on…
Reliable Large Language Models (LLMs) should abstain when confidence is insufficient. However, prior studies often treat refusal as a generic "I don't know'', failing to distinguish input-level ambiguity (data uncertainty) from capability…
Hallucination, the generation of factually incorrect information, remains a significant challenge for large language models (LLMs), especially in open-domain long-form generation. Existing approaches for detecting hallucination in long-form…
Existing large language models (LLMs) occasionally generate plausible yet factually incorrect responses, known as hallucinations. Two main approaches have been proposed to mitigate hallucinations: retrieval-augmented language models (RALMs)…
Existing techniques for training language models can be misaligned with the truth: if we train models with imitation learning, they may reproduce errors that humans make; if we train them to generate text that humans rate highly, they may…
Imitation learning practitioners have often noted that conditioning policies on previous actions leads to a dramatic divergence between "held out" error and performance of the learner in situ. Interactive approaches can provably address…
Multimodal Large Language Models (MLLMs) can enhance trustworthiness by aligning with human preferences. As human preference labeling is laborious, recent works employ evaluation models for assessing MLLMs' responses, using the model-based…
While large language models (LLMs) demonstrate strong capabilities across diverse user queries, they still suffer from hallucinations, often arising from knowledge misalignment between pre-training and fine-tuning. To address this…