Overview

Language model probing studies what models represent internally and what they can do, using controlled tasks to examine hidden activations, attention, logits, and generated behavior. Common approaches include training linear or nonlinear readouts on activations, comparing representations across layers and prompts, and testing causal effects with interventions such as activation patching or steering. Probes are used to investigate linguistic knowledge, reasoning, uncertainty, hallucination, safety, and model performance, and can support practical tools for monitoring, routing, and evaluation. A central open tension is that decodable information may not be used to produce an answer, while probe results can also reflect prompt structure, labels, or dataset artifacts rather than the intended capability.

Latest papers

CardsList

300 papers30 papers in the last four weeks, down 3% on the four weeks before. 0.2% of all new papers.

  1. The Geometry of Lies: Knowledge and Persona Directions for Detecting Deception in Large Language Models

    Nov 19, 2026Md. Abdullah Al Zami, Suraiya Shanaz, Mahdi Hasan +3Deception in Language ModelsLLM Honesty

  2. Language models represent more about emphasis than they can tell you

    Nov 19, 2026Aarne Talman, Stergios ChatzikyriakidisSpeech ProsodyLanguage Model Probing

  3. GateWatch: Low-Overhead LLM Monitoring with Boolean Circuits

    Nov 18, 2026Firas Spanioli, Yaniv DavidLLM Safety MonitoringLanguage Model Probing

  4. Do Probing and Steering Favor the Same Directions Within and Across Language Models?

    Nov 14, 2026Afshin Sadeghi, Daniel BeckerLanguage Model SteeringActivation Steering

  5. Detection Is Not Early Warning: Evaluating Token-Level Alarms on Language Model Failures

    Nov 12, 2026Marco De Negri, Luca Sartori, Lorenzo Baggi +5Language Model MonitoringLanguage Model Error Detection

  6. A Closer Look at Language Model Introspection: Probing Information and Reporting

    Nov 11, 2026Dustin Rubin, Vincenzo Timmel, Xufeng He +3Language Model IntrospectionLanguage Model Probing

  7. Interpretability Is Not Model-Specific

    Nov 9, 2026Marcus Armstrong, Navid Ayoobi, Pradham Mummaleti +2Interpretable MLMechanistic Interpretability

  8. GRAP: Gender Rubric-Aware Probing for Analyzing Gender Bias in LLMs

    Nov 9, 2026Seona Seo, Geonhee Ahn, Hyunsoo Cho +1Gender Bias in Language ModelsLanguage Model Bias Evaluation

  9. Linear Probes to Bayesian Manifolds: Geometry of Arithmetic in Large Language Models

    Nov 8, 2026Anshul Kumar, Neha Prasad, Manoj Kumar +2LLM InterpretabilityRepresentation Geometry in Language Models

  10. Pole or Vector? Encoders Capture Bangla Compound Verbs, but LLMs Read Them Literally

    Nov 7, 2026Md. Asaduzzaman Shuvo, Md Mehrab Karim Antor, Md. Adnanul Islam +4Bengali NLPLanguage Model Probing

  11. What Do Collusion Probes Read? Three Pitfalls in Evaluating Multi-Agent Collusion Monitors

    Nov 7, 2026Mainul Hasan Mohsin, Mohammad Shahidur RahmanAI Agent MonitoringMulti-Agent System Security

  12. Scout Before You Route: Predicting Task Success and Token Count from a Small LLM Prefill

    Nov 6, 2026Aristides Milios, Christopher PalLLM Output Length PredictionLLM Routing

  13. It Takes Two to Probe, Thousands to Retrieve: Putting Vision--Language Probes to the Retrieval Test

    Oct 29, 2026Ivana Beňová, Michal Gregor, Marián ŠimkoVLM EvaluationVision-Language Retrieval

  14. Value Probes: Learned Value Functions for Beam-Search Red Teaming

    Oct 28, 2026Jonathan F. Jao, Andrew Kyle LampinenLLM AuditingLanguage Model Probing

  15. Robust LLM-Generated Text Detection through Matched Conditional-Response Probing

    Oct 28, 2026Junfeng Hao, Wenxian Wang, Tao Leng +3AI-Generated Text DetectionRobustness of Machine-Generated Text Detection

  16. Scaling Laws for Safety Probes

    Oct 22, 2026Yuchen Wu, Jerry WeiLLM Safety MonitoringLanguage Model Probing

  17. Same Beliefs, Different Policies: Learning Reusable Representations for Early Stopping in LLM Reasoning

    Oct 21, 2026Jingwei Ni, Dmitry Abulkhanov, Vladislav Smirnov +5Early Stopping for Language Model ReasoningEarly Stopping

  18. How Language Models Organize Task Information

    Oct 20, 2026Kaiyang Wan, Yuxia WangLanguage Model ProbingLanguage Model Performance Prediction