The paper introduces Guided-AP, a method that improves attribution patching by using a few exact measurements to identify and correct its errors. On a single prompt, Guided-AP reduces error by up to 26% across three behaviors, while on many prompts it cuts average error on factual completion by 75% to 79%. The approach leverages shared error signs across components and tasks to efficiently guide verification and correction.
arXiv:2606. 09899v1 Announce Type: cross Abstract: A central goal of mechanistic interpretability is to identify which internal components causally drive a language model's behavior.
Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weig...
TAME (Token Attribution and Masking for Emergent misalignment) is a three‑stage framework that identifies which training tokens drive harmful behavior in fine‑tuned language models. It first scores tokens by how much fine‑tuning increases their likelihood, then characterizes patterns among high‑attribution tokens, and finally validates them by masking during training. Experiments on Llama and Qwen show that masking the top‑attribution tokens reduces emergent misalignment by 23‑ to 36‑fold, while random masking has no effect.
The paper investigates mechanistic interpretability, focusing on how automated circuit discovery is evaluated. It shows that the commonly used faithfulness objective can favor circuits that reproduce a model’s behavior poorly, creating an objective-level recovery gap. Experiments on four human-reference tasks and InterpBench reveal that many discovery methods misrank candidate circuits, and that restoring excluded signals can correct most of these misrankings without altering the circuits’ behavior.
The paper investigates intrinsic self‑correction, where a language model revises its own answer without new evidence. Across 29 open‑weight LLMs on BoolQ, GSM8K, and Corr2Cause, the study tracks how revisions change correctness, revealing that while some models improve significantly, others lose a notable fraction of correct answers. The authors compare three runtime strategies—keeping the initial answer, always accepting the revision, and selectively gating revisions—and find that the best approach depends on the model and task, suggesting that self‑correction should be treated as a revision policy rather than a uniformly beneficial second pass.
The paper introduces DEDUCE, a three‑stage framework that turns large language models into proactive error correctors by detecting input fact errors, devising correction strategies, and delivering reliable answers. It also presents MisFactQA, a dataset of factual errors, and new metrics for robustness evaluation. Experiments on TruthfulQA, FalseQA, and MisFactQA show significant gains in accuracy and error correction across Qwen, LLaMA, and Gemma models.
The paper investigates automated failure attribution (AFA) in large language model (LLM)-based agentic systems, where multiple agents collaborate on complex tasks. It finds that LLMs contain sufficient internal knowledge to attribute failures but often fail to output the correct step, revealing a knows‑but‑won’t‑say gap. To address this, the authors introduce grafting—a manual intervention that leverages latent model knowledge—showing it consistently outperforms direct prompting across several LLMs and benchmarks.
The paper investigates how memory systems can answer a current query correctly yet fail to retain distinctions needed for later updates. Using a paired‑history audit, the authors evaluate 24 history pairs across six synthetic mechanisms and two model backends, achieving perfect reveal accuracy on DeepSeek and high accuracy on GLM. Record‑level audits reveal specific failures in structured reveal memories and frontier late‑reference adequacy, and the authors test a label‑equivariant repair that only partially restores correctness.
arXiv:2606. 05486v1 Announce Type: cross Abstract: Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, while existing attribution methods are designed to explain observable outputs such as logits or generated tokens.
arXiv:2609.37616v1 Announce Type: new Abstract: Language models tend to agree with whatever a user asserts, and post-training increasingly targets this sycophancy so that models evaluate claims on th...
Large language models (LLMs) frequently produce confident yet factually incorrect responses when user inputs contain misleading premises, a phenomenon we attribute to fact perturbations in the input....
arXiv:2606. 09876v1 Announce Type: new Abstract: Large language models often express high confidence in answers that are wrong.