{"total":90,"rows":[{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2404.16710","title":"LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding","created":"2024-04-25","cited_by_count":307,"area":"inference"},{"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":"2403.17887","title":"The Unreasonable Ineffectiveness of the Deeper Layers","created":"2024-03-26","cited_by_count":226,"area":"lr"},{"id":"2308.09124","title":"Linearity of Relation Decoding in Transformer Language Models","created":"2023-08-17","cited_by_count":200,"area":null},{"id":"2402.16837","title":"Do Large Language Models Latently Perform Multi-Hop Reasoning?","created":"2024-02-26","cited_by_count":169,"area":"lr"},{"id":"2402.16363","title":"LLM Inference Unveiled: Survey and Roofline Model Insights","created":"2024-02-26","cited_by_count":152,"area":"inference"},{"id":"2307.09458","title":"Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla","created":"2023-07-18","cited_by_count":149,"area":null},{"id":"2402.11187","title":"LaCo: Large Language Model Pruning via Layer Collapse","created":"2024-02-17","cited_by_count":126,"area":null},{"id":"2311.04897","title":"Future Lens: Anticipating Subsequent Tokens from a Single Hidden State","created":"2023-11-08","cited_by_count":121,"area":null},{"id":"2407.02646","title":"A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models","created":"2024-07-02","cited_by_count":118,"area":"mechinterp"},{"id":"2402.09025","title":"SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks","created":"2024-02-14","cited_by_count":97,"area":null},{"id":"2305.15054","title":"A Mechanistic Interpretation of Arithmetic Reasoning in Language Models using Causal Mediation Analysis","created":"2023-05-24","cited_by_count":89,"area":"lr"},{"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":"2607.15495","title":"Verbalizable Representations Form a Global Workspace in Language Models","created":"2026-07-16","cited_by_count":75,"area":null},{"id":"2312.04916","title":"EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism","created":"2023-12-08","cited_by_count":69,"area":"inference"},{"id":"2402.09221","title":"Spectral Filters, Dark Signals, and Attention Sinks","created":"2024-02-14","cited_by_count":56,"area":"mechinterp"},{"id":"2310.11877","title":"The Curious Case of Hallucinatory (Un)answerability: Finding Truths in the Hidden States of Over-Confident Large Language Models","created":"2023-10-18","cited_by_count":55,"area":null},{"id":"2402.09733","title":"Do LLMs Know about Hallucination? An Empirical Investigation of LLM's Hidden States","created":"2024-02-15","cited_by_count":52,"area":null},{"id":"2407.15017","title":"Knowledge Mechanisms in Large Language Models: A Survey and Perspective","created":"2024-07-22","cited_by_count":51,"area":null},{"id":"2401.12874","title":"From Understanding to Utilization: A Survey on Explainability for Large Language Models","created":"2024-01-23","cited_by_count":48,"area":null},{"id":"2410.05864","title":"From Tokens to Words: On the Inner Lexicon of LLMs","created":"2024-10-08","cited_by_count":48,"area":null},{"id":"2403.08946","title":"Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era","created":"2024-03-13","cited_by_count":46,"area":null},{"id":"2305.12535","title":"Explaining How Transformers Use Context to Build Predictions","created":"2023-05-21","cited_by_count":43,"area":null},{"id":"2406.12094","title":"Who's asking? User personas and the mechanics of latent misalignment","created":"2024-06-17","cited_by_count":40,"area":"safety"},{"id":"2406.00877","title":"Evidence of Learned Look-Ahead in a Chess-Playing Neural Network","created":"2024-06-02","cited_by_count":34,"area":null},{"id":"2305.13417","title":"VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers","created":"2023-05-22","cited_by_count":32,"area":null},{"id":"2402.12865","title":"Backward Lens: Projecting Language Model Gradients into the Vocabulary Space","created":"2024-02-20","cited_by_count":32,"area":null},{"id":"2411.04430","title":"Towards Unifying Interpretability and Control: Evaluation via Intervention","created":"2024-11-07","cited_by_count":32,"area":"mechinterp"},{"id":"2312.09299","title":"Weight subcloning: direct initialization of transformers using larger pretrained ones","created":"2023-12-14","cited_by_count":30,"area":null},{"id":"2403.05846","title":"Diffusion Lens: Interpreting Text Encoders in Text-to-Image Pipelines","created":"2024-03-09","cited_by_count":30,"area":"diffusion"},{"id":"2305.01628","title":"The Benefits of Bad Advice: Autocontrastive Decoding across Model Layers","created":"2023-05-02","cited_by_count":29,"area":null},{"id":"2310.18969","title":"Analyzing Vision Transformers for Image Classification in Class Embedding Space","created":"2023-10-29","cited_by_count":26,"area":null},{"id":"2409.09951","title":"Optimal ablation for interpretability","created":"2024-09-16","cited_by_count":25,"area":null},{"id":"2405.12250","title":"Your Transformer is Secretly Linear","created":"2024-05-19","cited_by_count":23,"area":null},{"id":"2310.03686","title":"DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers","created":"2023-10-05","cited_by_count":22,"area":null},{"id":"2311.07772","title":"In-context Learning and Gradient Descent Revisited","created":"2023-11-13","cited_by_count":20,"area":null},{"id":"2506.05140","title":"AudioLens: A Closer Look at Auditory Attribute Perception of Large Audio-Language Models","created":"2025-06-05","cited_by_count":18,"area":null},{"id":"2402.13137","title":"The Hidden Space of Transformer Language Adapters","created":"2024-02-20","cited_by_count":17,"area":null},{"id":"2412.11965","title":"Inferring Functionality of Attention Heads from their Parameters","created":"2024-12-16","cited_by_count":17,"area":null},{"id":"2404.07004","title":"LM Transparency Tool: Interactive Tool for Analyzing Transformer Language Models","created":"2024-04-10","cited_by_count":16,"area":null},{"id":"2510.18871","title":"How Do LLMs Use Their Depth?","created":"2025-10-21","cited_by_count":16,"area":null},{"id":"2504.20168","title":"MICE for CATs: Model-Internal Confidence Estimation for Calibrating Agents with Tools","created":"2025-04-28","cited_by_count":15,"area":"agents"},{"id":"2502.15007","title":"LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers","created":"2025-02-20","cited_by_count":12,"area":null},{"id":"2410.01288","title":"Mitigating Copy Bias in In-Context Learning through Neuron Pruning","created":"2024-10-02","cited_by_count":11,"area":null},{"id":"2506.04179","title":"SkipGPT: Dynamic Layer Pruning Reinvented with Token Awareness and Module Decoupling","created":"2025-06-04","cited_by_count":11,"area":null},{"id":"2506.05451","title":"Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety","created":"2025-06-05","cited_by_count":10,"area":"safety"},{"id":"2509.13316","title":"Do Activation Verbalization Methods Convey Privileged Information?","created":"2025-09-16","cited_by_count":10,"area":"mechinterp"},{"id":"2601.18091","title":"From LLMs to LRMs: Rethinking Pruning for Reasoning-Centric Models","created":"2026-01-26","cited_by_count":10,"area":"lr"},{"id":"2605.06510","title":"Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models","created":"2026-05-07","cited_by_count":10,"area":null},{"id":"2506.03700","title":"AdaDecode: Accelerating LLM Decoding with Adaptive Layer Parallelism","created":"2025-06-04","cited_by_count":7,"area":"lr"},{"id":"2511.19265","title":"Unboxing the Black Box: Mechanistic Interpretability for Algorithmic Understanding of Neural Networks","created":"2025-11-24","cited_by_count":6,"area":"mechinterp"},{"id":"2309.01479","title":"Parameter and Computation Efficient Transfer Learning for Vision-Language Pre-trained Models","created":"2023-09-04","cited_by_count":5,"area":"mllm"},{"id":"2412.17836","title":"Look Ahead Text Understanding and LLM Stitching","created":"2024-12-16","cited_by_count":4,"area":null},{"id":"2505.23183","title":"Unsupervised Word-level Quality Estimation for Machine Translation Through the Lens of Annotators (Dis)agreement","created":"2025-05-29","cited_by_count":4,"area":null},{"id":"2604.15557","title":"Predicting Where Steering Vectors Succeed","created":"2026-04-16","cited_by_count":4,"area":"mechinterp"},{"id":"2406.02721","title":"Self-Control of LLM Behaviors by Compressing Suffix Gradient into Prefix Controller","created":"2024-06-04","cited_by_count":3,"area":null},{"id":"2408.12959","title":"Multimodal Contrastive In-Context Learning","created":"2024-08-23","cited_by_count":3,"area":null},{"id":"2501.11006","title":"GREEN-CODE: Learning to Optimize Energy Efficiency in LLM-based Code Generation","created":"2025-01-19","cited_by_count":3,"area":null},{"id":"2503.10542","title":"Language Models, Graph Searching, and Supervision Adulteration: When More Supervision is Less and How to Make More More","created":"2025-03-13","cited_by_count":3,"area":null},{"id":"2505.19606","title":"Languages in Whisper-Style Speech Encoders Align Both Phonetically and Semantically","created":"2025-05-26","cited_by_count":3,"area":null},{"id":"2605.09011","title":"A Geometric Perspective on Next-Token Prediction in Large Language Models: Three Emerging Phases","created":"2026-05-09","cited_by_count":3,"area":null},{"id":"2407.15516","title":"Attention Is All You Need But You Don't Need All Of It For Inference of Large Language Models","created":"2024-07-22","cited_by_count":2,"area":null},{"id":"2504.19327","title":"DeepInsert: Early Layer Bypass for Efficient and Performant Multimodal Understanding","created":"2025-04-27","cited_by_count":2,"area":null},{"id":"2402.13113","title":"When Only Time Will Tell: Interpreting How Transformers Process Local Ambiguities Through the Lens of Restart-Incrementality","created":"2024-02-20","cited_by_count":1,"area":null},{"id":"2404.06709","title":"CQIL: Inference Latency Optimization with Concurrent Computation of Quasi-Independent Layers","created":"2024-04-10","cited_by_count":1,"area":null},{"id":"2409.14091","title":"Normalized Narrow Jump To Conclusions: Normalized Narrow Shortcuts for Parameter Efficient Early Exit Transformer Prediction","created":"2024-09-21","cited_by_count":1,"area":"inference"},{"id":"2410.13077","title":"Tuning Language Models by Mixture-of-Depths Ensemble","created":"2024-10-16","cited_by_count":1,"area":null},{"id":"2411.07071","title":"Universal Response and Emergence of Induction in LLMs","created":"2024-11-11","cited_by_count":1,"area":null},{"id":"2505.11576","title":"Concept-Guided Interpretability via Neural Chunking","created":"2025-05-16","cited_by_count":1,"area":null},{"id":"2508.18227","title":"GM-Skip: Metric-Guided Transformer Block Skipping for Efficient Vision-Language Models","created":"2025-08-20","cited_by_count":1,"area":"lr"},{"id":"2603.13259","title":"How Transformers Reject Wrong Answers: Rotational Dynamics of Factual Constraint Processing","created":"2026-02-25","cited_by_count":1,"area":"mechinterp"},{"id":"2603.20406","title":"Thinking in Different Spaces: Domain-Specific Latent Geometry Survives Cross-Architecture Translation","created":"2026-03-20","cited_by_count":1,"area":"lr"},{"id":"2605.03258","title":"The Right Answer, the Wrong Direction: Why Transformers Fail at Counting and How to Fix It","created":"2026-05-05","cited_by_count":1,"area":null},{"id":"2605.16023","title":"Judge Circuits Explain Format-Induced Inconsistency in LLM-as-a-Judge","created":"2026-05-15","cited_by_count":1,"area":null},{"id":"2605.16893","title":"NGM: A Plug-and-Play Training-Free Memory Module for LLMs","created":"2026-05-16","cited_by_count":1,"area":null},{"id":"2608.10251","title":"Off-Axis, On Purpose: Where a Transformer Computes Concepts and Why it Does So","created":"2026-08-10","cited_by_count":1,"area":null},{"id":"2609.21383","title":"Prediction Dynamics in Depth-Recurrent Language Models","created":"2026-09-18","cited_by_count":1,"area":null},{"id":"2502.01803","title":"Discovering Chunks in Neural Embeddings for Interpretability","created":"2025-02-03","cited_by_count":0,"area":null},{"id":"2504.13984","title":"One Jump Is All You Need: Short-Cutting Transformers for Early Exit Prediction with One Jump to Fit All Exit Levels","created":"2025-04-18","cited_by_count":0,"area":"inference"},{"id":"2507.17618","title":"SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token","created":"2025-07-23","cited_by_count":0,"area":"inference"},{"id":"2512.14880","title":"Task Matrices: Linear Maps for Cross-Model Finetuning Transfer","created":"2025-12-16","cited_by_count":0,"area":null},{"id":"2512.23837","title":"Adversarial Lens: Exploiting Attention Layers to Generate Adversarial Examples for Evaluation","created":"2025-12-29","cited_by_count":0,"area":"mechinterp"},{"id":"2602.10350","title":"When Less Is More? Diagnosing ASR Predictions in Sardinian via Layer-Wise Decoding","created":"2026-02-10","cited_by_count":0,"area":null},{"id":"2607.00083","title":"Harnessing the Latent Space: From Steering Vectors to Model Calibrators for Control and Trust","created":"2026-06-30","cited_by_count":0,"area":"mechinterp"},{"id":"2607.21613","title":"The Hard Decision Layer: Evidence for Committed Inference in Transformers","created":"2026-05-25","cited_by_count":0,"area":null},{"id":"2608.10260","title":"Interpreting Language Model Hidden States at Scale","created":"2026-08-10","cited_by_count":0,"area":null},{"id":"2608.29706","title":"The Depth Flow of Token Representations Is Nonlinear and Does Not Descend Its Own Density","created":"2026-08-30","cited_by_count":0,"area":null},{"id":"2609.18823","title":"Using OCR Heads to Verbalize Image Semantics","created":"2026-09-16","cited_by_count":0,"area":null},{"id":"2609.29848","title":"Encoded but Not Decoded: Layer-Localized Evidence for a Three-Level Gap in LLM Syntax","created":"2026-09-24","cited_by_count":0,"area":null}],"next_cursor":null}