LLM popularity judgments align more closely with pretraining data exposure counts than with Wikipedia popularity, with stronger effects in pairwise comparisons and larger models.
Title resolution pending
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
citation-role summary
background 3
other 1
citation-polarity summary
representative citing papers
STAR-KV applies differentiable soft thresholding for per-head and per-block adaptive low-rank KV cache compression, combined with hybrid decomposition and low-rank-aware quantization, achieving up to 75% compression and 3.1x throughput gains.
CyberEvolver introduces a four-layer self-evolving agent architecture with trace-to-diagnosis and population beam search that raises seed agent success rates by 13.6% on CTF, exploitation, and penetration tasks across four LLMs.
TriMix dynamically fuses logits from three model sources to outperform baselines and Proxy Tuning on eight low-resource languages across four model families.
SPAGBias reveals that LLMs form nuanced gender associations with specific urban micro-spaces that exceed real-world distributions and produce failures in planning and descriptive tasks.
Empirical study of LLM brand recommendations across industries finds moderate concentration (mean Gini 0.28) and low cross-model agreement (41.6%) on top brands.
The Ghost Annotator framework applies conformal prediction and collaborative filtering representations to measure LLM divergence from human annotations across four models and datasets, revealing higher confidence in misaligned cases and consistent demographic misalignment.
A large-scale audit of 21 LLMs on OR-Bench, XSTest, ToxiGen and BOLD using composition adjustment reveals distinct conservative vs permissive safety strategies, unequal demographic protection, and post-training stability within model families.
GRASPrune removes 50% of parameters from LLaMA-2-7B via global gating and projected straight-through estimation, reaching 12.18 WikiText-2 perplexity and competitive zero-shot accuracy after four epochs on 512 calibration sequences.
Introduces a parameter-driven framework for data attribution in LLMs that enables negotiation among creators, users, and intermediaries to meet stakeholder goals within the data economy.
DCLM-Baseline dataset lets a 7B model reach 64% 5-shot MMLU accuracy after 2.6T tokens, beating prior open-data models by 6.6 points on MMLU with 40% less compute.
BloombergGPT is a 50B parameter LLM trained on a 708B token mixed financial and general dataset that outperforms prior models on financial benchmarks while preserving general LLM performance.
Atlas reaches over 42% accuracy on Natural Questions with only 64 examples, outperforming a 540B-parameter model by 3% with 50x fewer parameters.
GPT-NeoX-20B is a publicly released 20B parameter autoregressive language model trained on the Pile that shows strong gains in five-shot reasoning over similarly sized prior models.
The authors introduce Survey-aware Machine Learning (SaML) as a nine-step guideline that integrates survey design metadata throughout the ML lifecycle to enable valid population inference from complex health surveys.
InternLM2 is a new open-source LLM that outperforms prior versions on 30 benchmarks and long-context tasks through scaled pre-training to 32k tokens and a conditional online RLHF alignment strategy.
A survey reviewing benchmark data contamination in LLMs, its impact on evaluation, and alternative assessment approaches.
citing papers explorer
-
The Refusal--Compliance Tradeoff: A Large-Scale Safety Behavior Audit of Large Language Models
A large-scale audit of 21 LLMs on OR-Bench, XSTest, ToxiGen and BOLD using composition adjustment reveals distinct conservative vs permissive safety strategies, unequal demographic protection, and post-training stability within model families.
-
GRASPrune: Global Gating for Budgeted Structured Pruning of Large Language Models
GRASPrune removes 50% of parameters from LLaMA-2-7B via global gating and projected straight-through estimation, reaching 12.18 WikiText-2 perplexity and competitive zero-shot accuracy after four epochs on 512 calibration sequences.