Imagen de Portada

Beyond Perplexity: Multi-dimensional Safety Evaluation of LLM Compression

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Xu, Zhichao, Gupta, Ashim, Li, Tao, Bentham, Oliver, Srikumar, Vivek
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Found in Translation: Measuring Multilingual LLM Consistency as Simple as Translate then Evaluate
por: Gupta, Ashim, et al.
Publicado: (2025)

InfiniteScienceGym: An Unbounded, Procedurally-Generated Benchmark for Scientific Analysis
por: Bentham, Oliver, et al.
Publicado: (2026)

Test-Time Scaling with Repeated Sampling Improves Multilingual Text Generation
por: Gupta, Ashim, et al.
Publicado: (2025)

State Space Models are Strong Text Rerankers
por: Xu, Zhichao, et al.
Publicado: (2024)

Defragmenting Language Models: An Interpretability-based Approach for Vocabulary Expansion
por: Mehta, Maitrey, et al.
Publicado: (2026)

An Empirical Investigation of Matrix Factorization Methods for Pre-trained Transformers
por: Gupta, Ashim, et al.
Publicado: (2024)

Whispers of Doubt Amidst Echoes of Triumph in NLP Robustness
por: Gupta, Ashim, et al.
Publicado: (2023)

In-Context Example Ordering Guided by Label Distributions
por: Xu, Zhichao, et al.
Publicado: (2024)

LLM-Symbolic Integration for Robust Temporal Tabular Reasoning
por: Kulkarni, Atharv, et al.
Publicado: (2025)

Unequal Voices: How LLMs Construct Constrained Queer Narratives
por: Ghosal, Atreya, et al.
Publicado: (2025)

Distillation versus Contrastive Learning: How to Train Your Rerankers
por: Xu, Zhichao, et al.
Publicado: (2025)

Multi-dimensional Evaluation of Empathetic Dialog Responses
por: Xu, Zhichao, et al.
Publicado: (2024)

Reinforcing Code Generation: Improving Text-to-SQL with Execution-Based Learning
por: Kulkarni, Atharv, et al.
Publicado: (2025)

Understanding the Logic of Direct Preference Alignment through Logic
por: Richardson, Kyle, et al.
Publicado: (2024)

Promptly Predicting Structures: The Return of Inference
por: Mehta, Maitrey, et al.
Publicado: (2024)

Enhancing Question Answering on Charts Through Effective Pre-training Tasks
por: Gupta, Ashim, et al.
Publicado: (2024)

Rethinking Perplexity: Revealing the Impact of Input Length on Perplexity Evaluation in LLMs
por: Cheng, Letian, et al.
Publicado: (2026)

Chain-of-Thought Unfaithfulness as Disguised Accuracy
por: Bentham, Oliver, et al.
Publicado: (2024)

The Perplexity Paradox: Why Code Compresses Better Than Math in LLM Prompts
por: Johnson, Warren
Publicado: (2026)

Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models
por: Ankner, Zachary, et al.
Publicado: (2024)

LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
por: Xu, Zhichao, et al.
Publicado: (2026)

Rethinking On-policy Optimization for Query Augmentation
por: Xu, Zhichao, et al.
Publicado: (2025)

Beyond Perplexity: Character Distribution Signatures and the MDTA Benchmark for AI Text Detection
por: Narayanasamy, Priyadarshan, et al.
Publicado: (2026)

JudgeAgent: Beyond Static Benchmarks for Knowledge-Driven and Dynamic LLM Evaluation
por: Shi, Zhichao, et al.
Publicado: (2025)

Multi-Lingual Cyber Threat Detection in Tweets/X Using ML, DL, and LLM: A Comparative Analysis
por: Murad, Saydul Akbar, et al.
Publicado: (2025)

Beyond Natural Language Perplexity: Detecting Dead Code Poisoning in Code Generation Datasets
por: Tsai, Chi-Chien, et al.
Publicado: (2025)

Beyond Perplexity: Let the Reader Select Retrieval Summaries via Spectrum Projection Score
por: Hu, Zhanghao, et al.
Publicado: (2025)

Perplexity Trap: PLM-Based Retrievers Overrate Low Perplexity Documents
por: Wang, Haoyu, et al.
Publicado: (2025)

Low-Perplexity LLM-Generated Sequences and Where To Find Them
por: Wuhrmann, Arthur, et al.
Publicado: (2025)

Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning
por: Wu, Chao-Chung, et al.
Publicado: (2025)

RLShield: Practical Multi-Agent RL for Financial Cyber Defense with Attack-Surface MDPs and Real-Time Response Orchestration
por: Nayak, Srikumar
Publicado: (2026)

Beyond English and Evasion: A Human-Annotated Multi-Domain Benchmark for High-Stakes LLM Safety Evaluation in Chinese
por: Zaghouani, Wajdi, et al.
Publicado: (2026)

Beyond Perplexity: A Lightweight Benchmark for Knowledge Retention in Supervised Fine-Tuning
por: Shabgahi, Soheil Zibakhsh, et al.
Publicado: (2026)

On the Fallacy of Global Token Perplexity in Spoken Language Model Evaluation
por: Hsu, Chan-Jan, et al.
Publicado: (2026)

Beyond One-Size-Fits-All: Multi-Domain, Multi-Task Framework for Embedding Model Selection
por: Khetan, Vivek
Publicado: (2024)

Can Perplexity Reflect Large Language Model's Ability in Long Text Understanding?
por: Hu, Yutong, et al.
Publicado: (2024)

The Hidden Dimensions of LLM Alignment: A Multi-Dimensional Analysis of Orthogonal Safety Directions
por: Pan, Wenbo, et al.
Publicado: (2025)

Rectify Evaluation Preference: Improving LLMs' Critique on Math Reasoning via Perplexity-aware Reinforcement Learning
por: Tian, Changyuan, et al.
Publicado: (2025)

PatentScore: Multi-dimensional Evaluation of LLM-Generated Patent Claims
por: Yoo, Yongmin, et al.
Publicado: (2025)

A Perplexity and Menger Curvature-Based Approach for Similarity Evaluation of Large Language Models
por: Zhang, Yuantao, et al.
Publicado: (2025)