From Latents to Wires: Surgical Post-Editing on Large Language Models
cs.AI
Submitted: 2026-09-26
Updated: 2026-09-26
Terminology
Sources
- Refusal in Language Models Is Mediated by a Single Direction
- Compressed Sensing for Capability Localization in Large Language Models
- Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Gradient Routing: Masking Gradients to Localize Computation in Neural Networks
- Towards Automated Circuit Discovery for Mechanistic Interpretability
- Who's Harry Potter? Approximate Unlearning in LLMs
- Localizing Model Behavior with Path Patching
- Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models
- Measuring Massive Multitask Language Understanding
- Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
- There Is More to Refusal in Large Language Models than a Single Direction
- A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity
- The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
- Double-I Watermark: Protecting Model Copyright for LLM Fine-tuning
- Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks
- Removing Backdoor-Based Watermarks in Neural Networks with Limited Data
- LLM-Pruner: On the Structural Pruning of Large Language Models
- TOFU: A Task of Fictitious Unlearning for LLMs
- Locating and Editing Factual Associations in GPT
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection