LLM Interpretability

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

57 papers in the last four weeks, up 73% on the four weeks before. 0.6% of all new papers.

Jul 13Week of Sep 28

Latest papers 546

All topics

Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces
May 14, 2026Sanjoy Chowdhury, Dinesh ManochaNeural Representation Geometry LLM Interpretability
University of Maryland, College Park, USA
Exemplar Partitioning for Mechanistic Interpretability
May 14, 2026Jessica RumbelowRepresentation Learning LLM Interpretability
Leap Laboratories
Dynamics of the Transformer Residual Stream: Coupling Spectral Geometry to Network Topology
May 14, 2026Jesseba Fernando, Grigori GuitchountsDynamical Systems Transformer
Network Science Institute, Northeastern University · Flagship Pioneering
Polar probe linearly decodes semantic structures from LLMs
May 13, 2026Pablo J. Diego-Simón, Pierre Orhan, Emmanuel Chemla +2Representation Learning Linear Probing
LSCP, ENS, PSL, EHESS, CNRS, Paris, France · Paris Brain Institute, Paris, France · Earth Species Project, Berkeley, California, USA +1
Rethinking Layer Relevance in Large Language Models Beyond Cosine Similarity
May 13, 2026Cristian Hinostroza, Rodrigo Toro Icarte, Christ Devia +4LLM Pruning LLM Interpretability
1Pontificia Universidad Cat´olica de Chile · 2National Center for Artificial Intelligence (CENIA) · 3Universidad de Chile
Probing Persona-Dependent Preferences in Language Models
May 13, 2026Oscar Gilg, Pierre Beckmann, Daniel Paleka +1LLM Alignment LLM Interpretability
MATS · MATS, EPFL · ETH Zürich +1
Tracing Persona Vectors Through LLM Pretraining
May 13, 2026Viktor Moskvoretskii, Dominik Glandorf, Jorge Medina Moreira +2Language Model Pretraining LLM Interpretability
EPFL
Controlling Logical Collapse in LLMs via Algebraic Ontology Projection over F2
May 13, 2026Hisashi Miyashita, Mgnite IncLLM Interpretability Reasoning Consistency in Language Models
Mgnite Inc.
Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling
May 13, 2026Adil AminLLM Alignment LLM Interpretability
ZEHEN Labs
When Attention Closes: How LLMs Lose the Thread in Multi-Turn Interaction
May 13, 2026Vardhan Dongre, Joseph Hsieh, Viet Dac Lai +3Self-Attention LLM Interpretability
University of Illinois Urbana-Champaign · Adobe Research
Descriptive Collision in Sparse Autoencoder Auto-Interpretability: When One Explanation Describes Many Features
May 13, 2026Jordan F. McCannLLM Interpretability Sparse Autoencoders
Independent Researcher
Layer-wise Representation Dynamics: An Empirical Investigation Across Embedders and Base LLMs
May 12, 2026Jingzhou Jiang, Yi Yang, Kar Yan TamText Embeddings LLM Pruning
The Hong Kong University of Science and Technology
All Circuits Lead to Rome: Rethinking Functional Anisotropy in Circuit and Sheaf Discovery for LLMs
May 12, 2026Xi Chen, Mingyu Jin, Jingcheng Niu +7Circuit Discovery LLM Interpretability
University of Toronto · The Hong Kong University of Science and Technology (Guangzhou) · Rutgers University +2
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
May 12, 2026Eric Bigelow, Raphaël Sarfati, Daniel Wurgaft +5LLM Interpretability In-Context Learning
GKnow: Measuring the Entanglement of Gender Bias and Factual Gender
May 12, 2026Leonor Veloso, Hinrich SchützeGender Bias in Language Models LLM Interpretability
Center for Information and Language Processing, LMU Munich · Munich Center for Machine Learning (MCML)
Targeted Neuron Modulation via Contrastive Pair Search
May 12, 2026Sam Herring, Jake Naviasky, Karan MalhotraLanguage Model Steering LLM Interpretability
Nous Research
Do Language Models Encode Knowledge of Linguistic Constraint Violations?
May 12, 2026Hardy, Sebastian PadóLLM Interpretability Mechanistic Interpretability
IMS, University of Stuttgart, Stuttgart, Germany · Universitas Mikroskil, Medan, Indonesia.
Domain Restriction via Multi SAE Layer Transitions
May 12, 2026Elias Shaheen, Avi MendelsonLLM Interpretability OOD Detection
Technion – Israel Institute of Technology, Haifa, Israel.
Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models
May 12, 2026Boyi Deng, Xu Wang, Yaoning Wang +15LLM Interpretability Sparse Autoencoders
Qwen Team
Macro: Enhancing Multilingual Counterfactual Explanations through Alignment-as-Preference Optimization
May 12, 2026Yilong Wang, Qianli Wang, Bohao Chu +3Multilingual Language Models Counterfactual Explanations
1Technische Universität Berlin · 2German Research Center for Artificial Intelligence (DFKI) · University of Duisburg-Essen +5
Drop the Act: Probe-Filtered RL for Faithful Chain-of-Thought Reasoning
May 12, 2026Swapnil Parekh, Naman GoyalCoT Faithfulness Faithfulness of Language Model Explanations
Intuit
A Mechanistic Investigation of Supervised Fine Tuning
May 12, 2026Ruhaan ChopraSupervised Fine-Tuning LLM Alignment
Independent Researcher
Compositional Literary Primitives in Instruction-Tuned LLMs: Cross-Architectural SAE Features for Self, Style, and Affect
May 11, 2026Joao Paulo Cavalcante Presa, Savio Salvarino Teles de OliveiraLLM Interpretability Personality Modeling in Language Models
Federal University of Goias (UFG)
Temporal Preference Concepts and their Functions in a Large Language Model
May 11, 2026Ian Rios-Sialer, Shantanu Darveshi, Shuai Jiang +4LLM Interpretability Temporal Reasoning in Language Models
AISC San Francisco, USA · AISC Munich, Germany · SPAR, AISC Berlin, Germany
Instructions Shape Production of Language, not Processing
May 11, 2026Andreas Waldis, Leshem Choshen, Yufang Hou +1LLM Interpretability Instruction Following
Department of Linguistics, University of Tübingen · IBM Research, MIT, and MIT-IBM Watson AI Lab · MIT, and MIT-IBM Watson AI Lab +2
SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing
May 11, 2026Mingxu Zhang, Yuhan Li, Lujundong Li +3Molecular Optimization LLM Interpretability
The Hong Kong University of Science and Technology (Guangzhou) · Nanjing University of Aeronautics and Astronautics · The 63rd Research Institute, National University of Defense Technology, Nanjing
Intrinsic Guardrails: How Semantic Geometry of Personality Interacts with Emergent Misalignment in LLMs
May 11, 2026Krishak Aneja, Manas Mittal, Anmol Goel +2LLM Alignment LLM Interpretability
International Institute of Information Technology, Hyderabad · TU Darmstadt · Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi
SLASH the Sink: Sharpening Structural Attention Inside LLMs
May 11, 2026Yiming Liu, Bin Lu, Xinbing Wang +2LLM Reasoning with Graphs LLM Interpretability
Shanghai Jiao Tong University, Shanghai, China · Institute of Geographical Science and Natural Resources Research, Chinese Academy of Sciences, Beijing, China
The Truth Lies Somewhere in the Middle (of the Generated Tokens)
May 11, 2026Sophie L. Wang, Phillip Isola, Brian CheungLLM Interpretability Autoregressive Language Modeling
1MIT.
Cross-Family Universality of Behavioral Axes via Anchor-Projected Representations
May 11, 2026Su-Hyeon Kim, Yo-Sub HanLLM Interpretability Transfer Learning
Department of Artificial Intelligence Yonsei University Seoul, South Korea · Department of Computer Science Yonsei University Seoul, South Korea

May 14, 2026, cs.AIJ/K move · Enter open · S save

Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces

Sanjoy Chowdhury, Dinesh Manocha

University of Maryland, College Park, USA

Neural Representation Geometry LLM Interpretability CoT Reasoning CoT Compression LLM Reasoning Reasoning Trace Analysis

Open paper AskPDF arXiv

Language models often generate long chain-of-thought traces, but it remains unclear how much of this reasoning is necessary for preserving the final prediction. We study this through the lens of overcomplete reasoning traces: generated traces that contain more intermediate steps than are needed to support the model's answer. We define the minimal core as the smallest subset of steps that preserves either the final answer or predictive distribution, and introduce metrics for compression ratio, redundancy mass, step necessity, and necessity concentration. Across six deliberative reasoning benchmarks spanning arithmetic, competition mathematics, expert scientific reasoning, and commonsense multi-hop QA, we find substantial overcompleteness: on average, 46% of steps are removable under greedy minimal-core extraction while preserving the original answer in 86% of cases. We also find that predictive support is concentrated: the top three steps account for 65% of measured necessity mass on average. Beyond compression, minimal cores expose a cleaner geometry of reasoning: compared with full traces, they improve correct-incorrect trace separation by 11 points, reduce estimated intrinsic dimensionality by 34%, and transfer across model families with 85% off-diagonal answer retention. Theoretically, we establish existence of minimal sufficient subsets, local irreducibility guarantees for greedy elimination, and certificates of overcompleteness and sparse necessity. Together, these results suggest that full reasoning traces are often verbose and overcomplete, while minimal cores isolate the effective support underlying language-model predictions.

Momentum

Related topics

Latest papers 546