Hugging Face Trending Papers
The Missing Minimal Pair: Stereotype Evaluation in LLMs
Read the original on Hugging Face Trending Papers →The Flow has not summarised this story yet — read it at Hugging Face Trending Papers.
Related stories
arXiv Computation and Language
6d ago
The Missing Minimal Pair: Stereotype Evaluation in LLMs
arXiv:2610.08747v1 Announce Type: new Abstract: A common approach to measuring bias in Large Language Models is to compare the log-likelihoods of two contrastive stereotype sentences. We argue that s...
arXiv Computation and Language
Sep 16
Deconstructing Stereotypes: Scope-Conditioned Generation for Effective Multilingual Counterspeech
The paper introduces a scope‑conditioned generation framework that incorporates structured stereotype characteristics into prompts for large language models, aiming to improve the quality of counterspeech against online hate speech. The authors validate the method on a new, human‑curated dataset in English, Italian, and Spanish, showing significant gains over generic baselines in factuality, specificity, cogency, and effectiveness for both explicit and implicit stereotypes.
Hugging Face Trending Papers
Sep 15
Deconstructing Stereotypes: Scope-Conditioned Generation for Effective Multilingual Counterspeech
Counterspeech (CS) - direct responses that counter online Hate Speech (HS) using reasoning and alternative viewpoints - has emerged as an alternative to content removal. Current automatic CS generatio...
arXiv Machine Learning
Jul 31
STEREODISCO: Discovering Stereotypicality in LLMs
arXiv:2607. 27824v1 Announce Type: cross Abstract: LLMs encode, convey, and perpetuate stereotypes.
arXiv AI
Sep 10
Same Values, Different Languages? From Multilingual Probing to Steering LLMs Toward Chinese Social Values
arXiv:2609.08515v1 Announce Type: cross Abstract: As Large Language Models (LLMs) are increasingly integrated into human society, aligning them with pluralistic social values has become a critical pr...
arXiv Computation and Language
Aug 27
Apples to Apples? Towards Comparable Crosslingual Language Model Evaluation
The paper investigates how to fairly compare language models across languages, noting that current evaluation methods vary widely and lack empirical validation. By training controlled monolingual models on parallel data and testing multilingual LLMs, the authors find that many normalized metrics suffer from biases due to tokenization, encoding, and orthographic differences. Instead, they recommend using sentence‑level negative log‑likelihood over semantically equivalent sequences for more reliable cross‑lingual comparisons.