:: Library Catalog

Εξώφυλλο

Αποθηκεύτηκε σε:

Λεπτομέρειες βιβλιογραφικής εγγραφής
Κύριοι συγγραφείς:	Beaglehole, Daniel, Radhakrishnan, Adityanarayanan, Boix-Adserà, Enric, Belkin, Mikhail
Μορφή:	Preprint
Έκδοση:	2025
Θέματα:	Computation and Language Artificial Intelligence Machine Learning
Διαθέσιμο Online:	https://arxiv.org/abs/2502.03708
Ετικέτες:	Προσθήκη ετικέτας Δεν υπάρχουν, Καταχωρήστε ετικέτα πρώτοι!

Παρόμοια τεκμήρια

Towards a theory of model distillation
από: Boix-Adsera, Enric
Έκδοση: (2024)

Secret mixtures of experts inside your LLM
από: Boix-Adsera, Enric
Έκδοση: (2025)

On the inductive bias of infinite-depth ResNets and the bottleneck rank
από: Boix-Adsera, Enric
Έκδοση: (2025)

The Features at Convergence Theorem: a first-principles alternative to the Neural Feature Ansatz for how networks learn representations
από: Boix-Adsera, Enric, κ.ά.
Έκδοση: (2025)

xRFM: Accurate, scalable, and interpretable feature learning models for tabular data
από: Beaglehole, Daniel, κ.ά.
Έκδοση: (2025)

Let Me Think! A Long Chain-of-Thought Can Be Worth Exponentially Many Short Ones
από: Mirtaheri, Parsa, κ.ά.
Έκδοση: (2025)

Contextual Linear Activation Steering of Language Models
από: Hsu, Brandon, κ.ά.
Έκδοση: (2026)

The Weight Gram Matrix Captures Sequential Feature Linearization in Deep Networks
από: Cha, Taehun, κ.ά.
Έκδοση: (2026)

The power of fine-grained experts: Granularity boosts expressivity in Mixture of Experts
από: Boix-Adsera, Enric, κ.ά.
Έκδοση: (2025)

When can transformers reason with abstract symbols?
από: Boix-Adsera, Enric, κ.ά.
Έκδοση: (2023)

Emergence in non-neural models: grokking modular arithmetic via average gradient outer product
από: Mallinar, Neil, κ.ά.
Έκδοση: (2024)

Catching rationalization in the act: detecting motivated reasoning before and after CoT via activation probing
από: Mirtaheri, Parsa, κ.ά.
Έκδοση: (2026)

Linear Recursive Feature Machines provably recover low-rank matrices
από: Radhakrishnan, Adityanarayanan, κ.ά.
Έκδοση: (2024)

Quadratic models for understanding catapult dynamics of neural networks
από: Zhu, Libin, κ.ά.
Έκδοση: (2022)

Convergent Evolution: How Different Language Models Learn Similar Number Representations
από: Fu, Deqing, κ.ά.
Έκδοση: (2026)

Efficient semantic uncertainty quantification in language models via diversity-steered sampling
από: Park, Ji Won, κ.ά.
Έκδοση: (2025)

Context-Scaling versus Task-Scaling in In-Context Learning
από: Abedsoltan, Amirhesam, κ.ά.
Έκδοση: (2024)

Catapults in SGD: spikes in the training loss and their impact on generalization through feature learning
από: Zhu, Libin, κ.ά.
Έκδοση: (2023)

Can sparse autoencoders be used to decompose and interpret steering vectors?
από: Mayne, Harry, κ.ά.
Έκδοση: (2024)

Efficient and accurate steering of Large Language Models through attention-guided feature learning
από: Davarmanesh, Parmida, κ.ά.
Έκδοση: (2026)

Merge to Mix: Mixing Datasets via Model Merging
από: Tao, Zhixu Silvia, κ.ά.
Έκδοση: (2025)

The merged-staircase property: a necessary and nearly sufficient condition for SGD learning of sparse functions on two-layer neural networks
από: Abbe, Emmanuel, κ.ά.
Έκδοση: (2022)

Towards Conversational Diagnostic AI
από: Tu, Tao, κ.ά.
Έκδοση: (2024)

Domain-level metacognitive monitoring in frontier LLMs: A 33-model atlas
από: Cacioli, Jon-Paul
Έκδοση: (2026)

Towards Conversational AI for Disease Management
από: Palepu, Anil, κ.ά.
Έκδοση: (2025)

Towards Execution-Grounded Automated AI Research
από: Si, Chenglei, κ.ά.
Έκδοση: (2026)

ClimateGPT: Towards AI Synthesizing Interdisciplinary Research on Climate Change
από: Thulke, David, κ.ά.
Έκδοση: (2024)

Feature learning as alignment: a structural property of gradient descent in non-linear neural networks
από: Beaglehole, Daniel, κ.ά.
Έκδοση: (2024)

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
από: Lu, Chris, κ.ά.
Έκδοση: (2024)

Large Language Models in the Task of Automatic Validation of Text Classifier Predictions
από: Tsymbalov, Aleksandr, κ.ά.
Έκδοση: (2025)

RaguTeam at SemEval-2026 Task 8: Meno and Friends in a Judge-Orchestrated LLM Ensemble for Faithful Multi-Turn Response Generation
από: Bondarenko, Ivan, κ.ά.
Έκδοση: (2026)

AI-AI Bias: large language models favor communications generated by large language models
από: Laurito, Walter, κ.ά.
Έκδοση: (2024)

Towards Automated Patent Workflows: AI-Orchestrated Multi-Agent Framework for Intellectual Property Management and Analysis
από: Srinivas, Sakhinana Sagar, κ.ά.
Έκδοση: (2024)

Towards AI-$45^{\circ}$ Law: A Roadmap to Trustworthy AGI
από: Yang, Chao, κ.ά.
Έκδοση: (2024)

Towards physician-centered oversight of conversational diagnostic AI
από: Vedadi, Elahe, κ.ά.
Έκδοση: (2025)

Out-of-Distribution Detection using Synthetic Data Generation
από: Abbas, Momin, κ.ά.
Έκδοση: (2025)

Wikontic: Constructing Wikidata-Aligned, Ontology-Aware Knowledge Graphs with Large Language Models
από: Chepurova, Alla, κ.ά.
Έκδοση: (2025)

Limitations of Normalization in Attention Mechanism
από: Mudarisov, Timur, κ.ά.
Έκδοση: (2025)

Scaling Transformer to 1M tokens and beyond with RMT
από: Bulatov, Aydar, κ.ά.
Έκδοση: (2023)

Toward Super Agent System with Hybrid AI Routers
από: Yao, Yuhang, κ.ά.
Έκδοση: (2025)