{"total":1353,"rows":[{"id":"2306.03341","title":"Inference-Time Intervention: Eliciting Truthful Answers from a Language Model","created":"2023-06-06","cited_by_count":1302,"area":null},{"id":"2308.10248","title":"Steering Language Models With Activation Engineering","created":"2023-08-20","cited_by_count":922,"area":"mechinterp"},{"id":"2102.12452","title":"Probing Classifiers: Promises, Shortcomings, and Advances","created":"2021-02-24","cited_by_count":876,"area":null},{"id":"2311.03658","title":"The Linear Representation Hypothesis and the Geometry of Large Language Models","created":"2023-11-07","cited_by_count":716,"area":null},{"id":"2310.06824","title":"The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets","created":"2023-10-10","cited_by_count":712,"area":null},{"id":"2211.15661","title":"What learning algorithm is in-context learning? Investigations with linear models","created":"2022-11-28","cited_by_count":684,"area":null},{"id":"2108.08810","title":"Do Vision Transformers See Like Convolutional Neural Networks?","created":"2021-08-19","cited_by_count":675,"area":null},{"id":"2303.08112","title":"Eliciting Latent Predictions from Transformers with the Tuned Lens","created":"2023-03-14","cited_by_count":623,"area":"mechinterp"},{"id":"2111.15640","title":"Diffusion Autoencoders: Toward a Meaningful and Decodable Representation","created":"2021-11-30","cited_by_count":449,"area":"diffusion"},{"id":"2210.13382","title":"Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task","created":"2022-10-24","cited_by_count":431,"area":null},{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2305.01610","title":"Finding Neurons in a Haystack: Case Studies with Sparse Probing","created":"2023-05-02","cited_by_count":334,"area":null},{"id":"2310.02207","title":"Language Models Represent Space and Time","created":"2023-10-03","cited_by_count":316,"area":null},{"id":"2502.02013","title":"Layer by Layer: Uncovering Hidden Representations in Language Models","created":"2025-02-04","cited_by_count":311,"area":null},{"id":"2406.15927","title":"Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs","created":"2024-06-22","cited_by_count":259,"area":null},{"id":"2206.07568","title":"Contrastive Learning as Goal-Conditioned Reinforcement Learning","created":"2022-06-15","cited_by_count":253,"area":null},{"id":"2401.06102","title":"Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models","created":"2024-01-11","cited_by_count":235,"area":"mechinterp"},{"id":"2106.07682","title":"Revisiting Model Stitching to Compare Neural Representations","created":"2021-06-14","cited_by_count":209,"area":null},{"id":"2403.17834","title":"Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography","created":"2024-03-26","cited_by_count":199,"area":null},{"id":"2107.12808","title":"Open-Ended Learning Leads to Generally Capable Agents","created":"2021-07-27","cited_by_count":182,"area":null},{"id":"2104.06644","title":"Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little","created":"2021-04-14","cited_by_count":177,"area":null},{"id":"2409.14507","title":"A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders","created":"2024-09-22","cited_by_count":160,"area":"mechinterp"},{"id":"2307.15771","title":"The Hydra Effect: Emergent Self-repair in Language Model Computations","created":"2023-07-28","cited_by_count":156,"area":null},{"id":"2106.09667","title":"Poisoning and Backdooring Contrastive Learning","created":"2021-06-17","cited_by_count":151,"area":null},{"id":"2311.12424","title":"Looped Transformers are Better at Learning Learning Algorithms","created":"2023-11-21","cited_by_count":149,"area":null},{"id":"2410.20526","title":"Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders","created":"2024-10-27","cited_by_count":143,"area":"mechinterp"},{"id":"2106.09647","title":"Deep Learning Through the Lens of Example Difficulty","created":"2021-06-17","cited_by_count":141,"area":null},{"id":"2207.13243","title":"Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks","created":"2022-07-27","cited_by_count":129,"area":null},{"id":"2401.14446","title":"Black-Box Access is Insufficient for Rigorous AI Audits","created":"2024-01-25","cited_by_count":126,"area":null},{"id":"2111.09259","title":"Acquisition of Chess Knowledge in AlphaZero","created":"2021-11-17","cited_by_count":123,"area":null},{"id":"2303.00566","title":"Structured Pruning for Deep Convolutional Neural Networks: A survey","created":"2023-03-01","cited_by_count":122,"area":null},{"id":"2403.10516","title":"FeatUp: A Model-Agnostic Framework for Features at Any Resolution","created":"2024-03-15","cited_by_count":116,"area":null},{"id":"2502.16681","title":"Are Sparse Autoencoders Useful? A Case Study in Sparse Probing","created":"2025-02-23","cited_by_count":115,"area":"mechinterp"},{"id":"2111.00034","title":"Neural Networks as Kernel Learners: The Silent Alignment Effect","created":"2021-10-29","cited_by_count":106,"area":null},{"id":"2309.05858","title":"Uncovering mesa-optimization algorithms in Transformers","created":"2023-09-11","cited_by_count":100,"area":null},{"id":"2104.07762","title":"Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?","created":"2021-04-15","cited_by_count":93,"area":null},{"id":"2402.17811","title":"TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space","created":"2024-02-27","cited_by_count":91,"area":null},{"id":"2210.16156","title":"Reliability of CKA as a Similarity Measure in Deep Learning","created":"2022-10-28","cited_by_count":87,"area":null},{"id":"2112.09130","title":"Ensembling Off-the-shelf Models for GAN Training","created":"2021-12-16","cited_by_count":86,"area":null},{"id":"2101.06072","title":"Video Summarization Using Deep Neural Networks: A Survey","created":"2021-01-15","cited_by_count":85,"area":null},{"id":"2405.00208","title":"A Primer on the Inner Workings of Transformer-based Language Models","created":"2024-04-30","cited_by_count":85,"area":null},{"id":"2407.12831","title":"Truth is Universal: Robust Detection of Lies in LLMs","created":"2024-07-03","cited_by_count":79,"area":null},{"id":"2307.00175","title":"Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblocks","created":"2023-06-30","cited_by_count":78,"area":null},{"id":"2311.17901","title":"SODA: Bottleneck Diffusion Models for Representation Learning","created":"2023-11-29","cited_by_count":77,"area":"diffusion"},{"id":"2406.00034","title":"Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories","created":"2024-05-26","cited_by_count":77,"area":"mechinterp"},{"id":"2310.10616","title":"How Do Transformers Learn In-Context Beyond Simple Functions? A Case Study on Learning with Representations","created":"2023-10-16","cited_by_count":75,"area":null},{"id":"2607.15495","title":"Verbalizable Representations Form a Global Workspace in Language Models","created":"2026-07-16","cited_by_count":75,"area":null},{"id":"2211.01201","title":"Human alignment of neural network representations","created":"2022-11-02","cited_by_count":74,"area":null},{"id":"2102.13249","title":"Chess as a Testbed for Language Model State Tracking","created":"2021-02-26","cited_by_count":72,"area":null},{"id":"2312.17249","title":"Do Androids Know They're Only Dreaming of Electric Sheep?","created":"2023-12-28","cited_by_count":67,"area":"mechinterp"},{"id":"2404.12038","title":"Uncovering Safety Risks of Large Language Models through Concept Activation Vector","created":"2024-04-18","cited_by_count":67,"area":"alignment"},{"id":"2502.08640","title":"Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs","created":"2025-02-12","cited_by_count":67,"area":"agents"},{"id":"2101.11109","title":"First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT","created":"2021-01-26","cited_by_count":65,"area":null},{"id":"2107.05677","title":"Codified audio language modeling learns useful representations for music information retrieval","created":"2021-07-12","cited_by_count":64,"area":null},{"id":"2109.09234","title":"Conditional probing: measuring usable information beyond a baseline","created":"2021-09-19","cited_by_count":62,"area":null},{"id":"2312.01037","title":"Eliciting Latent Knowledge from Quirky Language Models","created":"2023-12-02","cited_by_count":62,"area":null},{"id":"2403.15498","title":"Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models","created":"2024-03-21","cited_by_count":62,"area":"wm"},{"id":"2204.08831","title":"Probing for the Usage of Grammatical Number","created":"2022-04-19","cited_by_count":59,"area":null},{"id":"2206.07835","title":"Disentangling visual and written concepts in CLIP","created":"2022-06-15","cited_by_count":58,"area":null},{"id":"2302.02672","title":"Identifiability of latent-variable and structural-equation models: from linear to nonlinear","created":"2023-02-06","cited_by_count":58,"area":null},{"id":"2310.05707","title":"Guiding Language Model Reasoning with Planning Tokens","created":"2023-10-09","cited_by_count":58,"area":"lr"},{"id":"2209.11222","title":"Concept Activation Regions: A Generalized Framework For Concept-Based Explanations","created":"2022-09-22","cited_by_count":57,"area":null},{"id":"2303.15180","title":"Detecting Backdoors in Pre-trained Encoders","created":"2023-03-23","cited_by_count":55,"area":"safety"},{"id":"2305.03689","title":"COLA: A Benchmark for Compositional Text-to-image Retrieval","created":"2023-05-05","cited_by_count":53,"area":"mllm"},{"id":"2506.10805","title":"Detecting High-Stakes Interactions with Activation Probes","created":"2025-06-12","cited_by_count":53,"area":null},{"id":"2212.01378","title":"ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning","created":"2022-12-02","cited_by_count":52,"area":null},{"id":"2412.02104","title":"Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey","created":"2024-12-03","cited_by_count":52,"area":"mllm"},{"id":"2412.08686","title":"LatentQA: Teaching LLMs to Decode Activations Into Natural Language","created":"2024-12-11","cited_by_count":51,"area":null},{"id":"2505.18235","title":"The Origins of Representation Manifolds in Large Language Models","created":"2025-05-23","cited_by_count":51,"area":"mechinterp"},{"id":"2211.11567","title":"Neural networks trained with SGD learn distributions of increasing complexity","created":"2022-11-21","cited_by_count":50,"area":null},{"id":"2402.11917","title":"A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task","created":"2024-02-19","cited_by_count":49,"area":null},{"id":"2303.08117","title":"Do Transformers Parse while Predicting the Masked Word?","created":"2023-03-14","cited_by_count":48,"area":null},{"id":"2310.17230","title":"Codebook Features: Sparse and Discrete Interpretability for Neural Networks","created":"2023-10-26","cited_by_count":48,"area":null},{"id":"2507.06952","title":"What Has a Foundation Model Found? Using Inductive Bias to Probe for World Models","created":"2025-07-09","cited_by_count":47,"area":"wm"},{"id":"2601.11516","title":"Building Production-Ready Probes For Gemini","created":"2026-01-16","cited_by_count":47,"area":"safety"},{"id":"2406.07882","title":"Designing a Dashboard for Transparency and Control of Conversational AI","created":"2024-06-12","cited_by_count":46,"area":null},{"id":"2410.14670","title":"Decomposing The Dark Matter of Sparse Autoencoders","created":"2024-10-18","cited_by_count":45,"area":"mechinterp"},{"id":"2503.18878","title":"I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders","created":"2025-03-24","cited_by_count":45,"area":"lr"},{"id":"2302.00785","title":"SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysis","created":"2023-02-01","cited_by_count":44,"area":null},{"id":"2303.03363","title":"Enhancing Activity Prediction Models in Drug Discovery with the Ability to Understand Human Language","created":"2023-03-06","cited_by_count":44,"area":null},{"id":"2502.00379","title":"Latent Action Learning Requires Supervision in the Presence of Distractors","created":"2025-02-01","cited_by_count":44,"area":null},{"id":"2403.20009","title":"On Large Language Models' Hallucination with Regard to Known Facts","created":"2024-03-29","cited_by_count":43,"area":null},{"id":"2406.10625","title":"On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models","created":"2024-06-15","cited_by_count":43,"area":"lr"},{"id":"2502.17598","title":"Hallucination Detection in LLMs Using Spectral Features of Attention Maps","created":"2025-02-24","cited_by_count":43,"area":null},{"id":"2106.14282","title":"A Closer Look at How Fine-tuning Changes BERT","created":"2021-06-27","cited_by_count":42,"area":null},{"id":"2401.16765","title":"A Cross-Language Investigation into Jailbreak Attacks in Large Language Models","created":"2024-01-30","cited_by_count":41,"area":"safety"},{"id":"2601.04603","title":"Constitutional Classifiers++: Efficient Production-Grade Defenses against Universal Jailbreaks","created":"2026-01-08","cited_by_count":41,"area":"safety"},{"id":"2406.12673","title":"Estimating Knowledge in Large Language Models Without Generating a Single Token","created":"2024-06-18","cited_by_count":40,"area":null},{"id":"2505.17006","title":"CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning","created":"2025-05-22","cited_by_count":40,"area":null},{"id":"2603.05488","title":"Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought","created":"2026-03-05","cited_by_count":40,"area":"lr"},{"id":"2105.14002","title":"What if This Modified That? Syntactic Interventions via Counterfactual Embeddings","created":"2021-05-28","cited_by_count":39,"area":null},{"id":"2505.14685","title":"Language Models use Lookbacks to Track Beliefs","created":"2025-05-20","cited_by_count":39,"area":null},{"id":"2404.05583","title":"Towards More General Video-based Deepfake Detection through Facial Component Guided Adaptation for Foundation Model","created":"2024-04-08","cited_by_count":38,"area":null},{"id":"2410.06166","title":"Temporal Reasoning Transfer from Text to Video","created":"2024-10-08","cited_by_count":38,"area":"lr"},{"id":"2205.14252","title":"Self-supervised models of audio effectively explain human cortical responses to speech","created":"2022-05-27","cited_by_count":37,"area":null},{"id":"2507.08802","title":"The Non-Linear Representation Dilemma: Is Causal Abstraction Enough for Mechanistic Interpretability?","created":"2025-07-11","cited_by_count":37,"area":"mechinterp"},{"id":"2108.11308","title":"What do pre-trained code models know about code?","created":"2021-08-25","cited_by_count":36,"area":null},{"id":"2210.17020","title":"A Law of Data Separation in Deep Learning","created":"2022-10-31","cited_by_count":36,"area":null},{"id":"2406.14673","title":"Insights into LLM Long-Context Failures: When Transformers Know but Don't Tell","created":"2024-06-20","cited_by_count":36,"area":null},{"id":"2301.05709","title":"Self-Supervised Image-to-Point Distillation via Semantically Tolerant Contrastive Loss","created":"2023-01-12","cited_by_count":35,"area":null}],"next_cursor":"100"}