Copertina

Task-Circuit Quantization: Leveraging Knowledge Localization and Interpretability for Compression

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Xiao, Hanqi, Sung, Yi-Lin, Stengel-Eskin, Elias, Bansal, Mohit
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Artificial Intelligence Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

UPCORE: Utility-Preserving Coreset Selection for Balanced Unlearning
di: Patil, Vaidehi, et al.
Pubblicazione: (2025)

Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind
di: Xiao, Hanqi, et al.
Pubblicazione: (2026)

ReGAL: Refactoring Programs to Discover Generalizable Abstractions
di: Stengel-Eskin, Elias, et al.
Pubblicazione: (2024)

Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Models
di: Prasad, Archiki, et al.
Pubblicazione: (2023)

Multi-Attribute Steering of Language Models via Targeted Intervention
di: Nguyen, Duy, et al.
Pubblicazione: (2025)

Soft Self-Consistency Improves Language Model Agents
di: Wang, Han, et al.
Pubblicazione: (2024)

DataEnvGym: Data Generation Agents in Teacher Environments with Student Feedback
di: Khan, Zaid, et al.
Pubblicazione: (2024)

Executable Functional Abstractions: Inferring Generative Programs for Advanced Math Problems
di: Khan, Zaid, et al.
Pubblicazione: (2025)

One Life to Learn: Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration
di: Khan, Zaid, et al.
Pubblicazione: (2025)

Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression
di: Singh, Joykirat, et al.
Pubblicazione: (2025)

Contrastive Region Guidance: Improving Grounding in Vision-Language Models without Training
di: Wan, David, et al.
Pubblicazione: (2024)

Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns
di: Xiao, Hanqi, et al.
Pubblicazione: (2025)

Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates
di: Nguyen, Duy, et al.
Pubblicazione: (2026)

Skill-Based Mixture-of-Experts: Adaptive Routing for Heterogeneous Reasoning via Inferred Skills
di: Chen, Justin Chih-Yao, et al.
Pubblicazione: (2025)

AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals
di: Nguyen, Duy, et al.
Pubblicazione: (2026)

Language Models Identify Ambiguities and Exploit Loopholes
di: Choi, Jio, et al.
Pubblicazione: (2025)

LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models
di: Stengel-Eskin, Elias, et al.
Pubblicazione: (2024)

Teaching Models to Balance Resisting and Accepting Persuasion
di: Stengel-Eskin, Elias, et al.
Pubblicazione: (2024)

PRInTS: Reward Modeling for Long-Horizon Information Seeking
di: Lee, Jaewoo, et al.
Pubblicazione: (2025)

System-1.x: Learning to Balance Fast and Slow Planning with Language Models
di: Saha, Swarnadeep, et al.
Pubblicazione: (2024)

Learning to Generate Unit Tests for Automated Debugging
di: Prasad, Archiki, et al.
Pubblicazione: (2025)

The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration
di: Patil, Vaidehi, et al.
Pubblicazione: (2025)

Cog-DRIFT: Exploration on Adaptively Reformulated Instances Enables Learning from Hard Reasoning Problems
di: Chen, Justin Chih-Yao, et al.
Pubblicazione: (2026)

RSQ: Learning from Important Tokens Leads to Better Quantized LLMs
di: Sung, Yi-Lin, et al.
Pubblicazione: (2025)

Retrieval-Augmented Generation with Conflicting Evidence
di: Wang, Han, et al.
Pubblicazione: (2025)

Are language models rational? The case of coherence norms and belief revision
di: Hofweber, Thomas, et al.
Pubblicazione: (2024)

GTBench: Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluations
di: Duan, Jinhao, et al.
Pubblicazione: (2024)

ComPEFT: Compression for Communicating Parameter Efficient Updates via Sparsification and Quantization
di: Yadav, Prateek, et al.
Pubblicazione: (2023)

LASeR: Learning to Adaptively Select Reward Models with Multi-Armed Bandits
di: Nguyen, Duy, et al.
Pubblicazione: (2024)

CAPTURe: Evaluating Spatial Reasoning in Vision Language Models via Occluded Object Counting
di: Pothiraj, Atin, et al.
Pubblicazione: (2025)

RotBench: Evaluating Multimodal Large Language Models on Identifying Image Rotation
di: Niu, Tianyi, et al.
Pubblicazione: (2025)

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs
di: Nguyen, Duy, et al.
Pubblicazione: (2025)

GenerationPrograms: Fine-grained Attribution with Executable Programs
di: Wan, David, et al.
Pubblicazione: (2025)

MAMM-Refine: A Recipe for Improving Faithfulness in Generation with Multi-Agent Collaboration
di: Wan, David, et al.
Pubblicazione: (2025)

Inducing Systematicity in Transformers by Attending to Structurally Quantized Embeddings
di: Jiang, Yichen, et al.
Pubblicazione: (2024)

Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs?
di: Hase, Peter, et al.
Pubblicazione: (2024)

Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy
di: Li, Pingzhi, et al.
Pubblicazione: (2023)

Routing with Generated Data: Annotation-Free LLM Skill Estimation and Expert Selection
di: Niu, Tianyi, et al.
Pubblicazione: (2026)

ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models
di: Sung, Yi-Lin, et al.
Pubblicazione: (2023)

See It from My Perspective: How Language Affects Cultural Bias in Image Understanding
di: Ananthram, Amith, et al.
Pubblicazione: (2024)