{"total":33,"rows":[{"id":"2607.15495","title":"Verbalizable Representations Form a Global Workspace in Language Models","created":"2026-07-16","cited_by_count":75,"area":null},{"id":"2605.26242","title":"Can LLMs Introspect? A Reality Check","created":"2026-05-25","cited_by_count":12,"area":null},{"id":"2604.16812","title":"Introspection Adapters: Training LLMs to Report Their Learned Behaviors","created":"2026-04-18","cited_by_count":11,"area":null},{"id":"2605.13329","title":"Tracing Persona Vectors Through LLM Pretraining","created":"2026-05-13","cited_by_count":7,"area":null},{"id":"2603.04426","title":"Delta-Crosscoder: Robust Crosscoder Model Diffing in Narrow Fine-Tuning Regimes","created":"2026-02-16","cited_by_count":4,"area":null},{"id":"2605.29601","title":"Training Deliberative Monitors for Black-Box Scheming Detection","created":"2026-05-28","cited_by_count":4,"area":null},{"id":"2606.32038","title":"Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision","created":"2026-06-30","cited_by_count":4,"area":null},{"id":"2604.22027","title":"Shared Lexical Task Representations Explain Behavioral Variability In LLMs","created":"2026-04-23","cited_by_count":2,"area":null},{"id":"2605.09314","title":"How LLMs Are Persuaded: A Few Attention Heads, Rerouted","created":"2026-05-10","cited_by_count":2,"area":null},{"id":"2604.07615","title":"ADAG: Automatically Describing Attribution Graphs","created":"2026-04-08","cited_by_count":1,"area":null},{"id":"2605.26045","title":"Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals","created":"2026-05-25","cited_by_count":1,"area":null},{"id":"2606.02609","title":"Building Better Activation Oracles","created":"2026-05-23","cited_by_count":1,"area":null},{"id":"2606.12268","title":"The Impossibility of Eliciting Latent Knowledge","created":"2026-06-10","cited_by_count":1,"area":null},{"id":"2606.26071","title":"Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment","created":"2026-06-24","cited_by_count":1,"area":null},{"id":"2607.14111","title":"Introspection Fine-Tuning (IFT): Training Small LLMs to Introspect","created":"2026-05-08","cited_by_count":1,"area":null},{"id":"2607.23379","title":"When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles","created":"2026-07-25","cited_by_count":1,"area":null},{"id":"2609.25518","title":"Matryoshka attribution: Learning to attribute language model outputs to representations and weights","created":"2026-09-22","cited_by_count":1,"area":null},{"id":"2602.10352","title":"Learning Self-Interpretation from Interpretability Artifacts: Training Lightweight Adapters on Vector-Label Pairs","created":"2026-02-10","cited_by_count":0,"area":null},{"id":"2603.20531","title":"Epistemic Observability in Language Models","created":"2026-03-20","cited_by_count":0,"area":null},{"id":"2606.09563","title":"PRISM: Recovering Instruction Sets from Language Model Activations","created":"2026-06-08","cited_by_count":0,"area":null},{"id":"2606.11445","title":"Forecasting Future Behavior as a Learning Task","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2607.04222","title":"Unsupervised Features Mining via Activation Geometry","created":"2026-07-05","cited_by_count":0,"area":null},{"id":"2607.16451","title":"Committed Before Reasoning: Behavioral Reproduction and Preliminary Activation-Level Evidence of Answer Pre-Commitment in an Open-Weight LLM","created":"2026-07-17","cited_by_count":0,"area":null},{"id":"2607.21774","title":"Probing Latent Colombian Identity Inferences in Qwen2.5-7B with Natural Language Autoencoders","created":"2026-07-23","cited_by_count":0,"area":null},{"id":"2607.27092","title":"Sky sphere representation in language models","created":"2026-07-29","cited_by_count":0,"area":null},{"id":"2609.24504","title":"On Emergent Capabilities and Model Merging","created":"2026-09-21","cited_by_count":0,"area":null},{"id":"2609.32890","title":"A Function-Level Vulnerability Score Measures Flag Rate More Than the Model: Protocol Effects on Paired Benchmarks","created":"2026-09-26","cited_by_count":0,"area":null},{"id":"2609.33702","title":"Understanding Confabulation and Rethinking Reconstruction in Activation Explanations","created":"2026-09-27","cited_by_count":0,"area":null},{"id":"2609.37040","title":"Selecting The Most Informative Tokens in Natural Language Autoencoders","created":"2026-09-29","cited_by_count":0,"area":null},{"id":"2609.37731","title":"Weights Read and Write Features: Scalable Parameter Decomposition Grounded in Activation Space","created":"2026-09-29","cited_by_count":0,"area":null},{"id":"2610.11012","title":"Distillation for Incrimination and Distillation for Capabilities","created":"2026-10-07","cited_by_count":0,"area":null},{"id":"2610.17495","title":"A Sense of Proprioception in Language Models","created":"2026-10-12","cited_by_count":0,"area":null},{"id":"2610.18727","title":"On the Monitorability of Natural Language Autoencoders: Identification vs Discrimination","created":"2026-10-13","cited_by_count":0,"area":null}],"next_cursor":null}