LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods

The rapid advancement of Large Language Models (LLMs) has driven their expanding application across various fields. One of the most promising applications is their role as evaluators based on natural language responses, referred to as ''LLMs-as-judges''. This framework has attracted growing attention from both academia and industry due to their excellent effectiveness, ability to generalize across tasks, and interpretability in the form of natural language. This paper presents a comprehensive survey of the LLMs-as-judges paradigm from five key perspectives: Functionality, Methodology, Applications, Meta-evaluation, and Limitations. We begin by providing a systematic definition of LLMs-as-Judges and introduce their functionality (Why use LLM judges?). Then we address methodology to construct an evaluation system with LLMs (How to use LLM judges?). Additionally, we investigate the potential domains for their application (Where to use LLM judges?) and discuss methods for evaluating them in various contexts (How to evaluate LLM judges?). Finally, we provide a detailed analysis of the limitations of LLM judges and discuss potential future directions. Through a structured and comprehensive analysis, we aim aims to provide insights on the development and application of LLMs-as-judges in both research and practice. We will continue to maintain the relevant resource list at https://github.com/CSHaitao/Awesome-LLMs-as-Judges.

A Collaborative Ensemble Framework for CTR Prediction

Recent advances in foundation models have established scaling laws that enable the development of larger models to achieve enhanced performance, motivating extensive research into large-scale recommendation models. However, simply increasing the model size in recommendation systems, even with large amounts of data, does not always result in the expected performance improvements. In this paper, we propose a novel framework, Collaborative Ensemble Training Network (CETNet), to leverage multiple distinct models, each with its own embedding table, to capture unique feature interaction patterns. Unlike naive model scaling, our approach emphasizes diversity and collaboration through collaborative learning, where models iteratively refine their predictions. To dynamically balance contributions from each model, we introduce a confidence-based fusion mechanism using general softmax, where model confidence is computed via negation entropy. This design ensures that more confident models have a greater influence on the final prediction while benefiting from the complementary strengths of other models. We validate our framework on three public datasets (AmazonElectronics, TaobaoAds, and KuaiVideo) as well as a large-scale industrial dataset from Meta, demonstrating its superior performance over individual models and state-of-the-art baselines. Additionally, we conduct further experiments on the Criteo and Avazu datasets to compare our method with the multi-embedding paradigm. Our results show that our framework achieves comparable or better performance with smaller embedding sizes, offering a scalable and efficient solution for CTR prediction tasks.

InterFormer: Towards Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction

Click-through rate (CTR) prediction, which predicts the probability of a user clicking an ad, is a fundamental task in recommender systems. The emergence of heterogeneous information, such as user profile and behavior sequences, depicts user interests from different aspects. A mutually beneficial integration of heterogeneous information is the cornerstone towards the success of CTR prediction. However, most of the existing methods suffer from two fundamental limitations, including (1) insufficient inter-mode interaction due to the unidirectional information flow between modes, and (2) aggressive information aggregation caused by early summarization, resulting in excessive information loss. To address the above limitations, we propose a novel module named InterFormer to learn heterogeneous information interaction in an interleaving style. To achieve better interaction learning, InterFormer enables bidirectional information flow for mutually beneficial learning across different modes. To avoid aggressive information aggregation, we retain complete information in each data mode and use a separate bridging arch for effective information selection and summarization. Our proposed InterFormer achieves state-of-the-art performance on three public datasets and a large-scale industrial dataset.

MA^2: A Self-Supervised and Motion Augmenting Autoencoder for Gait-Based Automatic Disease Detection

Ground reaction force (GRF) is the force exerted by the ground on a body in contact with it. GRF-based automatic disease detection (ADD) has become an emerging medical diagnosis method, which aims to learn and identify disease patterns corresponding to different gait pressures based on deep learning methods. Although existing ADD methods can save doctors time in making diagnoses, training deep models still struggles with the cost caused by the labeling engineering for a large number of gait diagnostic data for subjects. On the other hand, the accuracy of the deep model under the unified benchmark GRF dataset and the generalization ability on scalable gait datasets need to be further improved. To address these issues, we propose MA2, a GRF-based self-supervised and motion augmenting auto-encoder, which models the ADD task as an encoder-decoder paradigm. In the encoder, we introduce an embedding block including the 3-layer 1D convolution for extracting the token and a mask generator to randomly mask out the sequence of tokens to maximize the model's potential to capture high-level, discriminative, intrinsic representations. whereafter, the decoder utilizes this information to reconstruct the pixel sequence of the origin input and calculate the reconstruction loss to optimize the network. Moreover, the backbone of an auto-encoder is multi-head self-attention that can consider the global information of the token from the input, not just the local neighborhood. This allows the model to capture generalized contextual information. Extensive experiments demonstrate MA2 has SOTA performance of 90.91% accuracy on 1% limited pathological GRF samples with labels, and good generalization ability of 78.57% accuracy on scalable Parkinson disease dataset.

CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judges

The use of large language models (LLMs) as automated evaluation tools to assess the quality of generated natural language, known as LLMs-as-Judges, has demonstrated promising capabilities and is rapidly gaining widespread attention. However, when applied to pairwise comparisons of candidate responses, LLM-based evaluators often exhibit selection bias. Specifically, their judgments may become inconsistent when the option positions or ID tokens are swapped, compromising the effectiveness and fairness of the evaluation result. To address this challenge, we introduce CalibraEval, a novel label-free method for mitigating selection bias during inference. Specifically, CalibraEval reformulates debiasing as an optimization task aimed at adjusting observed prediction distributions to align with unbiased prediction distributions. To solve this optimization problem, we propose a non-parametric order-preserving algorithm (NOA). This algorithm leverages the partial order relationships between model prediction distributions, thereby eliminating the need for explicit labels and precise mathematical function modeling.Empirical evaluations of LLMs in multiple representative benchmarks demonstrate that CalibraEval effectively mitigates selection bias and improves performance compared to existing debiasing methods. This work marks a step toward building more robust and unbiased automated evaluation frameworks, paving the way for improved reliability in AI-driven assessments

An Automatic and Cost-Efficient Peer-Review Framework for Language Generation Evaluation

With the rapid development of large language models (LLMs), how to efficiently evaluate them has become an important research question. Existing evaluation methods often suffer from high costs, limited test formats, the need of human references, and systematic evaluation biases. To address these limitations, our study introduces the Auto-PRE, an automatic LLM evaluation framework based on peer review. In contrast to previous studies that rely on human annotations, Auto-PRE selects evaluator LLMs automatically based on their inherent traits including consistency, self-confidence, and pertinence. We conduct extensive experiments on three tasks: summary generation, non-factoid question-answering, and dialogue generation. Experimental results indicate our Auto-PRE achieves state-of-the-art performance at a lower cost. Moreover, our study highlights the impact of prompt strategies and evaluation formats on evaluation performance, offering guidance for method optimization in the future.

LexEval: A Comprehensive Chinese Legal Benchmark for Evaluating Large Language Models

Large language models (LLMs) have made significant progress in natural language processing tasks and demonstrate considerable potential in the legal domain. However, legal applications demand high standards of accuracy, reliability, and fairness. Applying existing LLMs to legal systems without careful evaluation of their potential and limitations could pose significant risks in legal practice. To this end, we introduce a standardized comprehensive Chinese legal benchmark LexEval. This benchmark is notable in the following three aspects: (1) Ability Modeling: We propose a new taxonomy of legal cognitive abilities to organize different tasks. (2) Scale: To our knowledge, LexEval is currently the largest Chinese legal evaluation dataset, comprising 23 tasks and 14,150 questions. (3) Data: we utilize formatted existing datasets, exam datasets and newly annotated datasets by legal experts to comprehensively evaluate the various capabilities of LLMs. LexEval not only focuses on the ability of LLMs to apply fundamental legal knowledge but also dedicates efforts to examining the ethical issues involved in their application. We evaluated 38 open-source and commercial LLMs and obtained some interesting findings. The experiments and findings offer valuable insights into the challenges and potential solutions for developing Chinese legal systems and LLM evaluation pipelines. The LexEval dataset and leaderboard are publicly available at \url{https://github.com/CSHaitao/LexEval} and will be continuously updated.

Hierarchical Structured Neural Network for Retrieval

Embedding Based Retrieval (EBR) is a crucial component of the retrieval stage in (Ads) Recommendation System that utilizes Two Tower or Siamese Networks to learn embeddings for both users and items (ads). It then employs an Approximate Nearest Neighbor Search (ANN) to efficiently retrieve the most relevant ads for a specific user. Despite the recent rise to popularity in the industry, they have a couple of limitations. Firstly, Two Tower model architecture uses a single dot product interaction which despite their efficiency fail to capture the data distribution in practice. Secondly, the centroid representation and cluster assignment, which are components of ANN, occur after the training process has been completed. As a result, they do not take into account the optimization criteria used for retrieval model. In this paper, we present Hierarchical Structured Neural Network (HSNN), a deployed jointly optimized hierarchical clustering and neural network model that can take advantage of sophisticated interactions and model architectures that are more common in the ranking stages while maintaining a sub-linear inference cost. We achieve 6.5% improvement in offline evaluation and also demonstrate 1.22% online gains through A/B experiments. HSNN has been successfully deployed into the Ads Recommendation system and is currently handling major portion of the traffic. The paper shares our experience in developing this system, dealing with challenges like freshness, volatility, cold start recommendations, cluster collapse and lessons deploying the model in a large scale retrieval production system.

LeKUBE: A Legal Knowledge Update BEnchmark

Recent advances in Large Language Models (LLMs) have significantly shaped the applications of AI in multiple fields, including the studies of legal intelligence. Trained on extensive legal texts, including statutes and legal documents, the legal LLMs can capture important legal knowledge/concepts effectively and provide important support for downstream legal applications such as legal consultancy. Yet, the dynamic nature of legal statutes and interpretations also poses new challenges to the use of LLMs in legal applications. Particularly, how to update the legal knowledge of LLMs effectively and efficiently has become an important research problem in practice. Existing benchmarks for evaluating knowledge update methods are mostly designed for the open domain and cannot address the specific challenges of the legal domain, such as the nuanced application of new legal knowledge, the complexity and lengthiness of legal regulations, and the intricate nature of legal reasoning. To address this gap, we introduce the Legal Knowledge Update BEnchmark, i.e. LeKUBE, which evaluates knowledge update methods for legal LLMs across five dimensions. Specifically, we categorize the needs of knowledge updates in the legal domain with the help of legal professionals, and then hire annotators from law schools to create synthetic updates to the Chinese Criminal and Civil Code as well as sets of questions of which the answers would change after the updates. Through a comprehensive evaluation of state-of-the-art knowledge update methods, we reveal a notable gap between existing knowledge update methods and the unique needs of the legal domain, emphasizing the need for further research and development of knowledge update mechanisms tailored for legal LLMs.

Mitigating Entity-Level Hallucination in Large Language Models

The emergence of Large Language Models (LLMs) has revolutionized how users access information, shifting from traditional search engines to direct question-and-answer interactions with LLMs. However, the widespread adoption of LLMs has revealed a significant challenge known as hallucination, wherein LLMs generate coherent yet factually inaccurate responses. This hallucination phenomenon has led to users' distrust in information retrieval systems based on LLMs. To tackle this challenge, this paper proposes Dynamic Retrieval Augmentation based on hallucination Detection (DRAD) as a novel method to detect and mitigate hallucinations in LLMs. DRAD improves upon traditional retrieval augmentation by dynamically adapting the retrieval process based on real-time hallucination detection. It features two main components: Real-time Hallucination Detection (RHD) for identifying potential hallucinations without external models, and Self-correction based on External Knowledge (SEK) for correcting these errors using external knowledge. Experiment results show that DRAD demonstrates superior performance in both detecting and mitigating hallucinations in LLMs. All of our code and data are open-sourced at https://github.com/oneal2000/EntityHallucination.