paper-with-me

Papers

How much do LLMs learn from negative examples?

2025-03-18 · Shadi Hamdan, Deniz Yuret

Large language models (LLMs) undergo a three-phase training process: unsupervised pre-training, supervised fine-tuning (SFT), and learning from human feedback (RLHF/DPO). Notably, it is during the final phase that these models are exposed to negative examples -- incorrect, rejected, or suboptimal responses to queries. This paper delves into the role of negative examples in the training of LLMs, using a likelihood-ratio (Likra) model on multiple-choice question answering benchmarks to precisely manage the influence and the volume of negative examples. Our findings reveal three key insights: (1) During a critical phase in training, Likra with negative examples demonstrates a significantly larger improvement per training example compared to SFT using only positive examples. This leads to a sharp jump in the learning curve for Likra unlike the smooth and gradual improvement of SFT; (2) negative examples that are plausible but incorrect (near-misses) exert a greater influence; and (3) while training with positive examples fails to significantly decrease the likelihood of plausible but incorrect answers, training with negative examples more accurately identifies them. These results indicate a potentially significant role for negative examples in improving accuracy and reducing hallucinations for LLMs.

📄 PDF Abstract BibTeX arXiv:2503.14391

Code (1)

shamdan17/likra 공식 구현 pytorch

Tasks

Multiple-choiceQuestion AnsweringUnsupervised Pre-training

Methods 이 논문이 사용한 방법론

SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…

Similar Papers 제목 키워드 기반

Large Language Model Unlearning

2023-10-14 · Yuanshun Yao, Xiaojun Xu, Yang Liu

We study how to perform unlearning, i.e. forgetting undesirable misbehaviors, on large language models (LLMs). We show at least three scenarios of aligning LLMs with human preferences can benefit from unlearning: (1) rem…

Language ModelingLanguage ModellingLarge Language Modelmodel+1

Hard negative examples are hard, but useful

2020-07-24 · ECCV 2020 8 · Hong Xuan, Abby Stylianou, Xiaotong Liu, Robert Pless

Triplet loss is an extremely common approach to distance metric learning. Representations of images from the same class are optimized to be mapped closer together in an embedding space than representations of images from…

Image RetrievalMetric LearningRetrievalSemantic Similarity+2

Defending Adversarial Examples by Negative Correlation Ensemble

2022-06-11 · Wenjian Luo, Hongwei Zhang, Linghao Kong, Zhijian Chen 외

The security issues in DNNs, such as adversarial examples, have attracted much attention. Adversarial examples refer to the examples which are capable to induce the DNNs return completely predictions by introducing caref…

Adversarial Robustness

Conan-embedding: General Text Embedding with More and Better Negative Samples

2024-08-28 · Shiyu Li, Yang Tang, ShiZhe Chen, Xi Chen

With the growing popularity of RAG, the capabilities of embedding models are gaining increasing attention. Embedding models are primarily trained through contrastive loss learning, with negative examples being a key comp…

Contrastive LearningGPURAG

GAPX: Generalized Autoregressive Paraphrase-Identification X

2022-10-05 · Yifei Zhou, Renyu Li, Hayden Housen, Ser-Nam Lim

Paraphrase Identification is a fundamental task in Natural Language Processing. While much progress has been made in the field, the performance of many state-of-the-art models often suffer from distribution shift during …

Paraphrase Identification