Copertina

The Dead Salmons of AI Interpretability

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Méloux, Maxime, Dirupo, Giada, Portet, François, Peyrard, Maxime
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Mechanistic Interpretability as Statistical Estimation: A Variance Analysis
di: Méloux, Maxime, et al.
Pubblicazione: (2025)

Everything, Everywhere, All at Once: Is Mechanistic Interpretability Identifiable?
di: Méloux, Maxime, et al.
Pubblicazione: (2025)

Meta-Statistical Learning: Supervised Learning of Statistical Estimators
di: Peyrard, Maxime, et al.
Pubblicazione: (2025)

Agentic AI: The Era of Semantic Decoding
di: Peyrard, Maxime, et al.
Pubblicazione: (2024)

Date Fragments: A Hidden Bottleneck of Tokenization for Temporal Reasoning
di: Bhatia, Gagan, et al.
Pubblicazione: (2025)

MIST: Mutual Information Estimation Via Supervised Training
di: Gritsai, German, et al.
Pubblicazione: (2025)

What Makes an LLM a Good Optimizer? A Trajectory Analysis of LLM-Guided Evolutionary Search
di: Zhang, Xinhao, et al.
Pubblicazione: (2026)

Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning
di: Geng, Saibo, et al.
Pubblicazione: (2023)

What Really Controls Temporal Reasoning in Large Language Models: Tokenisation or Representation of Time?
di: Bhatia, Gagan, et al.
Pubblicazione: (2026)

Interactive Evaluation of Large Language Models for Multi-Requirement Software Engineering Tasks
di: Rontogiannis, Dimitrios, et al.
Pubblicazione: (2025)

Novel-WD: Exploring acquisition of Novel World Knowledge in LLMs Using Prefix-Tuning
di: Méloux, Maxime, et al.
Pubblicazione: (2024)

Symbolic Autoencoding for Self-Supervised Sequence Learning
di: Amani, Mohammad Hossein, et al.
Pubblicazione: (2024)

Can GPT models Follow Human Summarization Guidelines? A Study for Targeted Communication Goals
di: Zhou, Yongxin, et al.
Pubblicazione: (2023)

Protein Secondary Structure Prediction Using Transformers
di: Maxime, Manzi Kevin
Pubblicazione: (2025)

Evaluating Language Model Agency through Negotiations
di: Davidson, Tim R., et al.
Pubblicazione: (2024)

Broken by Default: A Formal Verification Study of Security Vulnerabilities in AI-Generated Code
di: Blain, Dominik, et al.
Pubblicazione: (2026)

Flows: Building Blocks of Reasoning and Collaborating AI
di: Josifoski, Martin, et al.
Pubblicazione: (2023)

MedMeta: A Benchmark for LLMs in Synthesizing Meta-Analysis Conclusion from Medical Studies
di: Ha, Huy Hoang, et al.
Pubblicazione: (2026)

Improving LLM-Generated Code Quality with GRPO
di: Robeyns, Maxime, et al.
Pubblicazione: (2025)

Transformer-based Models to Deal with Heterogeneous Environments in Human Activity Recognition
di: EK, Sannara, et al.
Pubblicazione: (2022)

Can AI be Consentful?
di: Pistilli, Giada, et al.
Pubblicazione: (2025)

A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia
di: Monea, Giovanni, et al.
Pubblicazione: (2023)

OptiHive: Ensemble Selection for LLM-Based Optimization via Statistical Modeling
di: Bouscary, Maxime, et al.
Pubblicazione: (2025)

A Self-Improving Coding Agent
di: Robeyns, Maxime, et al.
Pubblicazione: (2025)

Building spatial world models from sparse transitional episodic memories
di: He, Zizhan, et al.
Pubblicazione: (2025)

Budget-Efficient Automatic Algorithm Design via Code Graph
di: Bouscary, Maxime, et al.
Pubblicazione: (2026)

RecSys Challenge 2023: From data preparation to prediction, a simple, efficient, robust and scalable solution
di: Manderlier, Maxime, et al.
Pubblicazione: (2024)

MuDreamer: Learning Predictive World Models without Reconstruction
di: Burchi, Maxime, et al.
Pubblicazione: (2024)

Helix-mRNA: A Hybrid Foundation Model For Full Sequence mRNA Therapeutics
di: Wood, Matthew, et al.
Pubblicazione: (2025)

Robustness Analysis of AI Models in Critical Energy Systems
di: Dogoulis, Pantelis, et al.
Pubblicazione: (2024)

LLM-as-a-Judge: Toward World Models for Slate Recommendation Systems
di: Bonin, Baptiste, et al.
Pubblicazione: (2025)

KCLNet: Physics-Informed Power Flow Prediction via Constraints Projections
di: Dogoulis, Pantelis, et al.
Pubblicazione: (2025)

RefAgent: A Multi-agent LLM-based Framework for Automatic Software Refactoring
di: Oueslati, Khouloud, et al.
Pubblicazione: (2025)

Distilling Genomic Models for Efficient mRNA Representation Learning via Embedding Matching
di: Haidari, Rasched, et al.
Pubblicazione: (2026)

Fully Autonomous AI Agents Should Not be Developed
di: Mitchell, Margaret, et al.
Pubblicazione: (2025)

PowerGraph-LLM: Novel Power Grid Graph Embedding and Optimization with Large Language Models
di: Bernier, Fabien, et al.
Pubblicazione: (2025)

Learning Transformer-based World Models with Contrastive Predictive Coding
di: Burchi, Maxime, et al.
Pubblicazione: (2025)

Accurate and Efficient World Modeling with Masked Latent Transformers
di: Burchi, Maxime, et al.
Pubblicazione: (2025)

The Sandbox Configurator: A Framework to Support Technical Assessment in AI Regulatory Sandboxes
di: Buscemi, Alessio, et al.
Pubblicazione: (2025)

Exploring Multimodal Foundation AI and Expert-in-the-Loop for Sustainable Management of Wild Salmon Fisheries in Indigenous Rivers
di: Xu, Chi, et al.
Pubblicazione: (2025)