paper-with-me

Papers

Aligning Model Evaluations with Human Preferences: Mitigating Token Count Bias in Language Model Assessments

2024-07-05 · Roland Daynauth, Jason Mars

The SLAM paper demonstrated that on-device Small Language Models (SLMs) are a viable and cost-effective alternative to API-based Large Language Models (LLMs), such as OpenAI's GPT-4, offering comparable performance and stability. However, SLAM also identified discrepancies between human preferences and traditional auto-evaluators. This follow-up paper explores methods to align LLM evaluator preferences with human evaluations by addressing biases, particularly toward higher token counts. We employed Bayesian statistics and a t-test to quantify this bias and developed a recalibration procedure to adjust the GPTScorer. Our findings significantly improve aligning the recalibrated LLM evaluator with human evaluations across multiple use cases. For instance, spearman's ranking correlation score in the Recommendation use case improved from -27.27 to 44.55. These results highlight the importance of accounting for biases in automated evaluations to ensure fair and accurate model assessments. The recalibration process enhances the reliability of automated evaluators, leading to better AI models that align with human values and expectations. This study provides a robust methodology for future research into bias correction and emphasizes the feasibility and benefits of developing human-aligned AI evaluation systems.

📄 PDF Abstract BibTeX arXiv:2407.12847

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage Modellingmodel

Methods 이 논문이 사용한 방법론

Attention 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Position-Wise Feed-Forward Layer 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

SpeechAlign: Aligning Speech Generation to Human Preferences

2024-04-08 · Dong Zhang, Zhaowei Li, ShiMin Li, Xin Zhang 외

Speech language models have significantly advanced in generating realistic speech, with neural codec language models standing out. However, the integration of human feedback to align speech outputs to human preferences i…

Language ModelingLanguage Modelling

Energy-Based Reward Models for Robust Language Model Alignment

2025-04-17 · Anamika Lochab, Ruqi Zhang

Reward models (RMs) are essential for aligning Large Language Models (LLMs) with human preferences. However, they often struggle with capturing complex human preferences and generalizing to unseen data. To address these …

Language ModelingLanguage Modelling

Token Preference Optimization with Self-Calibrated Visual-Anchored Rewards for Hallucination Mitigation

2024-12-19 · Jihao Gu, Yingyao Wang, Meng Cao, Pi Bu 외

Direct Preference Optimization (DPO) has been demonstrated to be highly effective in mitigating hallucinations in Large Vision Language Models (LVLMs) by aligning their outputs more closely with human preferences. Despit…

Hallucination

CHARM: Calibrating Reward Models With Chatbot Arena Scores

2025-04-14 · Xiao Zhu, Chenmien Tan, Pinzhen Chen, Rico Sennrich 외

Reward models (RMs) play a crucial role in Reinforcement Learning from Human Feedback by serving as proxies for human preferences in aligning large language models. In this paper, we identify a model preference bias in R…

Chatbot

Diffusion-RPO: Aligning Diffusion Models through Relative Preference Optimization

2024-06-10 · Yi Gu, Zhendong Wang, Yueqin Yin, Yujia Xie 외

Aligning large language models with human preferences has emerged as a critical focus in language modeling research. Yet, integrating preference learning into Text-to-Image (T2I) generative models is still relatively unc…

Language ModelingLanguage Modelling