Image de couverture de livre

Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Puyin, Li, Tan, Jiyuan, Jabbar, Ahmad, Icard, Thomas, Geiger, Atticus
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

How Causal Abstraction Underpins Computational Explanation
par: Geiger, Atticus, et autres
Publié: (2025)

A Reply to Makelov et al. (2023)'s "Interpretability Illusion" Arguments
par: Wu, Zhengxuan, et autres
Publié: (2024)

Modeling Discrimination with Causal Abstraction
par: Mossé, Milan, et autres
Publié: (2025)

Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretability
par: Geiger, Atticus, et autres
Publié: (2023)

Interpretability at Scale: Identifying Causal Mechanisms in Alpaca
par: Wu, Zhengxuan, et autres
Publié: (2023)

Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations
par: Geiger, Atticus, et autres
Publié: (2023)

How Do Transformers Learn Variable Binding in Symbolic Programs?
par: Wu, Yiwei, et autres
Publié: (2025)

Combining Causal Models for More Accurate Abstractions of Neural Networks
par: Pîslar, Theodora-Mara, et autres
Publié: (2025)

Are language models aware of the road not taken? Token-level uncertainty and hidden state dynamics
par: Zur, Amir, et autres
Publié: (2025)

Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution Behaviors
par: Huang, Jing, et autres
Publié: (2025)

Comparing Bad Apples to Good Oranges: Aligning Large Language Models via Joint Preference Optimization
par: Bansal, Hritik, et autres
Publié: (2024)

Updating CLIP to Prefer Descriptions Over Captions
par: Zur, Amir, et autres
Publié: (2024)

HyperSteer: Activation Steering at Scale with Hypernetworks
par: Sun, Jiuding, et autres
Publié: (2025)

Causal Abstraction in Model Interpretability: A Compact Survey
par: Zhang, Yihao
Publié: (2024)

Evaluating Open-Source Sparse Autoencoders on Disentangling Factual Knowledge in GPT-2 Small
par: Chaudhary, Maheep, et autres
Publié: (2024)

Generation Space Size: Understanding and Calibrating Open-Endedness of LLM Generations
par: Yu, Sunny, et autres
Publié: (2025)

From Noise to Signal: When Outliers Seed New Topics
par: Zve, Evangelia, et autres
Publié: (2026)

Fixing the Broken Compass: Diagnosing and Improving Inference-Time Reward Modeling
par: Li, Jiachun, et autres
Publié: (2025)

HyperDAS: Towards Automating Mechanistic Interpretability with Hypernetworks
par: Sun, Jiuding, et autres
Publié: (2025)

ReFT: Representation Finetuning for Language Models
par: Wu, Zhengxuan, et autres
Publié: (2024)

Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
par: Bigelow, Eric, et autres
Publié: (2026)

Reliable News or Propagandist News? A Neurosymbolic Model Using Genre, Topic, and Persuasion Techniques to Improve Robustness in Classification
par: Faye, Géraud, et autres
Publié: (2026)

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought
par: Boppana, Siddharth, et autres
Publié: (2026)

Notation Matters: A Benchmark Study of Token-Optimized Formats in Agentic AI Systems
par: Kutschka, Lorenz, et autres
Publié: (2026)

AgentCollabBench: Diagnosing When Good Agents Make Bad Collaborators
par: Mazumder, Aritra, et autres
Publié: (2026)

Measuring Embedding Sensitivity to Authorial Style in French: Comparing Literary Texts with Language Model Rewritings
par: Icard, Benjamin, et autres
Publié: (2026)

Embedding Style Beyond Topics: Analyzing Dispersion Effects Across Different Language Models
par: Icard, Benjamin, et autres
Publié: (2025)

On Probabilistic and Causal Reasoning with Summation Operators
par: Ibeling, Duligur, et autres
Publié: (2024)

Latent Abstraction for Retrieval-Augmented Generation
par: T, Ha Lan N., et autres
Publié: (2026)

Concept-Reversed Winograd Schema Challenge: Evaluating and Improving Robust Reasoning in Large Language Models via Abstraction
par: Han, Kaiqiao, et autres
Publié: (2024)

PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction
par: Yu, Simon, et autres
Publié: (2025)

From Scores to Steps: Diagnosing and Improving LLM Performance in Evidence-Based Medical Calculations
par: Wang, Benlu, et autres
Publié: (2025)

Belief in the Machine: Investigating Epistemological Blind Spots of Language Models
par: Suzgun, Mirac, et autres
Publié: (2024)

HYBRINFOX at CheckThat! 2024 -- Task 1: Enhancing Language Models with Structured Information for Check-Worthiness Estimation
par: Faye, Géraud, et autres
Publié: (2024)

HYBRINFOX at CheckThat! 2024 -- Task 2: Enriching BERT Models with the Expert System VAGO for Subjectivity Detection
par: Casanova, Morgane, et autres
Publié: (2024)

Detection vs. Execution: Single-Bucket Probes Miss Half the Mamba-2 State Sink
par: Jiang, Yuhang
Publié: (2026)

Abstraction-of-Thought Makes Language Models Better Reasoners
par: Hong, Ruixin, et autres
Publié: (2024)

A Causal Language Modeling Detour Improves Encoder Continued Pretraining
par: Touchent, Rian, et autres
Publié: (2026)

LLMs and the Abstraction and Reasoning Corpus: Successes, Failures, and the Importance of Object-based Representations
par: Xu, Yudong, et autres
Publié: (2023)

Enhancing Complex Causality Extraction via Improved Subtask Interaction and Knowledge Fusion
par: Gao, Jinglong, et autres
Publié: (2024)