LLM-as-Judge

Type inference for dynamic languages like Python is a persistent challenge in software engineering. While large language models (LLMs) have shown promise in code understanding, their type inference capabilities remain underexplored. We introduce `TypyBench`, a benchmark designed to evaluate LLMs' type inference across entire Python repositories. `TypyBench` features two novel metrics: `TypeSim`, which captures nuanced semantic relationships between predicted and ground truth types, and `TypeCheck`, which assesses type consistency across codebases. Our evaluation of various LLMs on a curated dataset of 50 high-quality Python repositories reveals that, although LLMs achieve decent `TypeSim` scores, they struggle with complex nested types and exhibit significant type consistency errors. These findings suggest that future research should shift focus from improving type similarity to addressing repository-level consistency. `TypyBench` provides a foundation for this new direction, offering insights into model performance across different type complexities and usage contexts. Our code and data are available at \href{https://github.com/typybench/typybench}.

Many existing evaluation benchmarks for Large Language Models (LLMs) quickly become outdated due to the emergence of new models and training data. These benchmarks also fall short in assessing how LLM performance changes over time, as they consist of a static set of questions without a temporal dimension. To address these limitations, we propose using future event prediction as a continuous evaluation method to assess LLMs' temporal generalization and forecasting abilities. Our benchmark, Daily Oracle, automatically generates question-answer (QA) pairs from daily news, challenging LLMs to predict "future" event outcomes. Our findings reveal that as pre-training data becomes outdated, LLM performance degrades over time. While Retrieval Augmented Generation (RAG) has the potential to enhance prediction accuracy, the performance degradation pattern persists, highlighting the need for continuous model updates. Code and data are available at https://agenticlearning.ai/daily-oracle.

Large language models are susceptible to memorizing repeated sequences, posing privacy and copyright concerns. A popular mitigation strategy is to remove memorized information from specific neurons post-hoc. However, such approaches have shown limited success so far. In a controlled setting, we show that the memorization of *natural* sequences (those that resemble linguistically plausible text) become *mechanistically entangled* with general language abilities, thereby becoming challenging to remove post-hoc. In this work, we put forward a new paradigm of MemSinks that promotes isolation of memorization by design. We leverage a sequence identifier to activate a unique set of memorization neurons for each sequence across repetitions. By analyzing the dynamics of learning and forgetting, we argue that MemSinks facilitates clean isolation of memorized content, making it easier to remove without compromising general language capabilities. We implement MemSinks at the billion-parameter and billion-token scale, and observe both effective isolation and strong generalization. To our knowledge, this is the first proof-of-concept on real data demonstrating that simultaneous generalization and isolation is achievable. We open-source our code at http://github.com/grghosal/MemSinks.

Post-training compression reduces the computational and memory costs of large language models (LLMs), enabling resource-efficient deployment. However, existing compression benchmarks focus narrowly on language modeling (e.g., perplexity) and natural language understanding tasks (e.g., GLUE accuracy), ignoring the agentic capabilities—workflow, tool use/function call, long-context understanding and real-world application. We introduce the Agent Compression Benchmark (ACBench), the first comprehensive benchmark for evaluating how compression impacts LLMs' agentic abilities. ACBench spans (1) 12 tasks across 4 capabilities (e.g., WorfBench for workflow generation, Needle-in-Haystack for long-context retrieval), (2) 4-bit quantization (GPTQ, AWQ) and 50% pruning (Wanda, SparseGPT), and (3) 15 models, including small (Gemma-2B), standard (Qwen2.5-7B), and distilled reasoning LLMs (DeepSeek-R1-Distill). Our experiments reveal compression tradeoffs: 4-bit quantization preserves workflow generation and tool use (1%--3% drop) but degrades real-world application accuracy by 10%--15%. We introduce ERank, Top-k Ranking Correlation and Energy to systematize analysis. ACBench provides actionable insights for optimizing LLM compression in agentic scenarios, bridging the gap between algorithmic efficiency and real-world applicability.

We investigate the use of large language models (LLMs) to simulate human responses to survey questions, and perform uncertainty quantification to gain reliable insights. Our approach converts imperfect, LLM-simulated responses into confidence sets for population parameters of human responses, addressing the distribution shift between the simulated and real populations. A key innovation lies in determining the optimal number of simulated responses: too many produce overly narrow confidence sets with poor coverage, while too few yield excessively loose estimates. To resolve this, our method adaptively selects the simulation sample size, ensuring valid average-case coverage guarantees. It is broadly applicable to any LLM, irrespective of its fidelity, and any procedure for constructing confidence sets. Additionally, the selected sample size quantifies the degree of misalignment between the LLM and the target human population. We illustrate our method on real datasets and LLMs.

Loss reweighting has shown significant benefits for machine unlearning with large language models (LLMs). However, their exact functionalities are left unclear and the optimal strategy remains an open question, thus impeding the understanding and improvement of existing methodologies. In this paper, we identify two distinct goals of loss reweighting, namely, Saturation and Importance---the former indicates that those insufficiently optimized data should be emphasized, while the latter stresses some critical data that are most influential for loss minimization. To study their usefulness, we design specific reweighting strategies for each goal and evaluate their respective effects on unlearning. We conduct extensive empirical analyses on well-established benchmarks, and summarize some important observations as follows: (i) Saturation enhances efficacy more than importance-based reweighting, and their combination can yield additional improvements. (ii) Saturation typically allocates lower weights to data with lower likelihoods, whereas importance-based reweighting does the opposite. (iii) The efficacy of unlearning is also largely influenced by the smoothness and granularity of the weight distributions. Based on these findings, we propose SatImp, a simple reweighting method that combines the advantages of both saturation and importance. Empirical results on extensive datasets validate the efficacy of our method, potentially bridging existing research gaps and indicating directions for future research. Our code is available at https://github.com/tmlr-group/SatImp.

Open-ended learning agents must efficiently prioritize goals in vast possibility spaces, focusing on those that maximize learning progress (LP). When such autotelic exploration is achieved by LLM agents trained with online RL in high-dimensional and evolving goal spaces, a key challenge for LP prediction is modeling one’s own competence, a form of metacognitive monitoring. Traditional approaches either require extensive sampling or rely on brittle expert-defined goal groupings. We introduce MAGELLAN, a metacognitive framework that lets LLM agents learn to predict their competence and learning progress online. By capturing semantic relationships between goals, MAGELLAN enables sample-efficient LP estimation and dynamic adaptation to evolving goal spaces through generalization. In an interactive learning environment, we show that MAGELLAN improves LP prediction efficiency and goal prioritization, being the only method allowing the agent to fully master a large and evolving goal space. These results demonstrate how augmenting LLM agents with a metacognitive ability for LP predictions can effectively scale curriculum learning to open-ended goal spaces.

It has become increasingly challenging to understand and evaluate LLM capabilities as these models exhibit a broader range of behaviors. In this position paper, we argue that LLM researchers should draw on the lessons from another field which has developed a rich set of experimental paradigms and design practices for probing the behavior of complex intelligent systems: animal cognition. We present five core principles of evaluation drawn from animal cognition research, and explain how they provide invaluable guidance for understanding LLM capabilities and behavior. We ground these principles in an empirical case study, and show how they can already provide a richer picture of one particular reasoning capability: transitive inference.

Evaluating Large Language Models (LLMs) often requires costly human annotations. To address this, LLM-based judges have been proposed, which compare the outputs of two LLMs enabling the ranking of models without human intervention. While several approaches have been proposed, many confounding factors are present between different papers. For instance the model, the prompt and other hyperparameters are typically changed at the same time making apple-to-apple comparisons challenging. In this paper, we propose to systematically analyze and tune the hyperparameters of LLM judges. To alleviate the high cost of evaluating a judge, we propose to leverage multi-objective multi-fidelity which allows to find judges that trades accuracy for cost and also reduce significantly the cost of the search. Our method identifies judges that not only outperform existing benchmarks in accuracy and cost-efficiency but also utilize open-weight models, ensuring greater accessibility and reproducibility.

Automatic evaluation methods based on large language models (LLMs) are emerging as the standard tool for assessing the instruction-following abilities of LLM-based agents. The most common method in this paradigm, pairwise comparisons with a baseline model, critically depends on the assumption of transitive preferences. However, the validity of this assumption remains largely unexplored. In this study, we investigate the presence of non-transitivity within the AlpacaEval framework and analyze its effects on model rankings. We find that LLM judges exhibit non-transitive preferences, leading to rankings that are sensitive to the choice of the baseline model. To mitigate this issue, we show that round-robin tournaments combined with Bradley-Terry models of preference can produce more reliable rankings. Notably, our method increases both the Spearman correlation and the Kendall correlation with Chatbot Arena (95.0\% $\rightarrow$ 96.4\% and 82.1\% $\rightarrow$ 86.3\% respectively). To address the computational cost of round-robin tournaments, we propose Swiss-Wise Iterative Matchmaking (Swim) tournaments, using a dynamic matching strategy to capture the benefits of round-robin tournaments while maintaining computational efficiency.

Accurate and verifiable large language model (LLM) simulations of human research subjects promise an accessible data source for understanding human behavior and training new AI systems. However, results to date have been limited, and few social scientists have adopted this method. In this position paper, we argue that the promise of LLM social simulations can be achieved by addressing five tractable challenges. We ground our argument in a review of empirical comparisons between LLMs and human research subjects, commentaries on the topic, and related work. We identify promising directions, including context-rich prompting and fine-tuning with social science datasets. We believe that LLM social simulations can already be used for pilot and exploratory studies, and more widespread use may soon be possible with rapidly advancing LLM capabilities. Researchers should prioritize developing conceptual models and iterative evaluations to make the best use of new AI systems.

One-shot post-training pruning enhances the deployment of billion-scale large language models (LLMs), with the pruning metric playing a pivotal role in determining which weights to remove. However, existing metrics underperform due to their reliance on a simple symbolic combination of weights and activations, overlooking imbalanced weight magnitudes and the disproportionate influence of activation outliers. To overcome these limitations, we introduce \textbf{BaWA}, a novel pruning metric that systematically \underline{Ba}lances \underline{W}eight and \underline{A}ctivation distributions for more effective pruning. BaWA introduces two key innovations: \textbf{magnitude normalization}, which mitigates weight imbalance across channels for fairer pruning decisions, and \textbf{outlier regularization}, which reduces the impact of activation outliers, ensuring more appropriate channel prioritization. To further enhance its effectiveness, BaWA incorporates an efficient and automatic framework for optimizing normalization and regularization hyperparameters. Extensive experiments validate BaWA as a state-of-the-art (SOTA) pruning metric. For instance, applying BaWA to induce 2:4 sparsity in Mistral-7B reduces perplexity in language comprehension by 2.49 and improves average downstream task accuracy by 3.08\%, outperforming the previous SOTA method Wanda.

Rigorous statistical evaluations of large language models (LLMs), including valid error bars and significance testing, are essential for meaningful and reliable performance assessment. Currently, when such statistical measures are reported, they typically rely on the Central Limit Theorem (CLT). In this position paper, we argue that while CLT-based methods for uncertainty quantification are appropriate when benchmarks consist of thousands of examples, they fail to provide adequate uncertainty estimates for LLM evaluations that rely on smaller, highly specialized benchmarks. In these small-data settings, we demonstrate that CLT-based methods perform very poorly, usually dramatically underestimating uncertainty (i.e. producing error bars that are too small). We give recommendations for alternative frequentist and Bayesian methods that are both easy to implement and more appropriate in these increasingly common scenarios.

As LLMs continue to become more powerful and versatile, human evaluation has become intractable at scale and reliance on automatic metrics has become the norm. Recently, it has been shown that LLMs are themselves state-of-the-art evaluators for many tasks. These *Autoraters* are typically designed so that they generalize to new systems *and* test sets. In practice, however, evaluation is performed on a small set of fixed, canonical test sets, which are carefully curated to measure the capabilities of interest and are not changed frequently. In this work, we design a method which specializes a prompted Autorater to a given test set, by leveraging historical ratings on the test set to construct in-context learning (ICL) examples. We evaluate our *Specialist* method on the task of fine-grained machine translation evaluation, and show that it dramatically outperforms the state-of-the-art XCOMET metric by 54% and 119% on the WMT'23 and WMT'24 test sets, respectively. We perform extensive analyses to understand the representations learned by our Specialist metrics, and how variability in rater behavior affects their performance. We also verify the generalizability and robustness of our Specialist method across different numbers of ICL examples, LLM backbones, systems to evaluate, and evaluation tasks.

Benchmark Data Contamination (BDC)—the inclusion of benchmark testing samples in the training set—has raised increasing concerns in Large Language Model (LLM) evaluation, leading to falsely inflated performance estimates and undermining evaluation reliability. To address this, researchers have proposed various mitigation strategies to update existing benchmarks, including modifying original questions or generating new ones based on them. However, a rigorous examination of the effectiveness of these mitigation strategies remains lacking. In this paper, we design a systematic and controlled pipeline along with two novel metrics—*fidelity* and *contamination resistance*—to provide a fine-grained and comprehensive assessment of existing BDC mitigation strategies. Previous assessment methods, such as accuracy drop and accuracy matching, focus solely on aggregate accuracy, often leading to incomplete or misleading conclusions. Our metrics address this limitation by emphasizing *question-level* evaluation result matching. Extensive experiments with 10 LLMs, 5 benchmarks, 20 BDC mitigation strategies, and 2 contamination scenarios reveal that no existing strategy effectively balances fidelity and contamination resistance. No semantic-preserving strategy yields a significant improvement in resistance over the vanilla case (*i.e.*, no benchmark update) across *all* benchmarks, while semantic-altering strategies sacrifice fidelity for resistance. These findings underscore the urgent need for designing more effective BDC mitigation strategies. Our code repository is available at https://github.com/ASTRAL-Group/BDC_mitigation_assessment.

Accurately predicting 3D structures and dynamics of physical systems is crucial in scientific applications. Existing approaches that rely on geometric Graph Neural Networks (GNNs) effectively enforce $\mathrm{E}(3)$-equivariance, but they often fail in leveraging extensive broader information. While direct application of Large Language Models (LLMs) can incorporate external knowledge, they lack the capability for spatial reasoning with guaranteed equivariance. In this paper, we propose EquiLLM, a novel framework for representing 3D physical systems that seamlessly integrates $\mathrm{E}(3)$-equivariance with LLM capabilities. Specifically, EquiLLM comprises four key components: geometry-aware prompting, an equivariant encoder, an LLM, and an equivariant adapter. Essentially, the LLM guided by the instructive prompt serves as a sophisticated invariant feature processor, while 3D directional information is exclusively handled by the equivariant encoder and adapter modules. Experimental results demonstrate that EquiLLM delivers significant improvements over previous methods across molecular dynamics simulation, human motion simulation, and antibody design, highlighting its promising generalizability.

Jailbreaking techniques trick Large Language Models (LLMs) into producing restricted output, posing a potential threat. One line of defense is to use another LLM as a Judge to evaluate the harmfulness of generated text. However, we reveal that these Judge LLMs are vulnerable to token segmentation bias, an issue that arises when delimiters alter the tokenization process, splitting words into smaller sub-tokens. This alters the embeddings of the entire sequence, reducing detection accuracy and allowing harmful content to be misclassified as safe. In this paper, we introduce Emoji Attack, a novel strategy that amplifies existing jailbreak prompts by exploiting token segmentation bias. Our method leverages in-context learning to systematically insert emojis into text before it is evaluated by a Judge LLM, inducing embedding distortions that significantly lower the likelihood of detecting unsafe content. Unlike traditional delimiters, emojis also introduce semantic ambiguity, making them particularly effective in this attack. Through experiments on state-of-the-art Judge LLMs, we demonstrate that Emoji Attack substantially reduces the unsafe prediction rate, bypassing existing safeguards.

LLM-as-a-Judge models generate chain-of-thought (CoT) sequences intended to capture the step-by-step reasoning process that underlies the final evaluation of a response. However, due to the lack of human-annotated CoTs for evaluation, the required components and structure of effective reasoning traces remain understudied. Consequently, previous approaches often (1) constrain reasoning traces to hand-designed components, such as a list of criteria, reference answers, or verification questions and (2) structure them such that planning is intertwined with the reasoning for evaluation. In this work, we propose EvalPlanner, a preference optimization algorithm for Thinking-LLM-as-a-Judge that first generates an unconstrained evaluation plan, followed by its execution, and then the final judgment. In a self-training loop, EvalPlanner iteratively optimizes over synthetically constructed evaluation plans and executions, leading to better final verdicts. Our method achieves a new state-of-the-art performance for generative reward models on RewardBench and PPE, despite being trained on fewer amount of, and synthetically generated, preference pairs. Additional experiments on other benchmarks like RM-Bench, JudgeBench, and FollowBenchEval further highlight the utility of both planning and reasoning for building robust LLM-as-a-Judge reasoning models.

Contemporary evaluation techniques are inadequate for agentic systems. These approaches either focus exclusively on final outcomes---ignoring the step-by-step nature of the thinking done by agentic systems---or require excessive manual labour. To address this, we introduce the **Agent-as-a-Judge** framework, wherein agentic systems are used to evaluate agentic systems. This is a natural extension of the LLM-as-a-Judge framework, incorporating agentic features that enable intermediate feedback for the entire task-solving processes for more precise evaluations. We apply the Agent-as-a-Judge framework to the task of code generation. To overcome issues with existing benchmarks and provide a proof-of-concept testbed for Agent-as-a-Judge, we present **DevAI**, a new benchmark of 55 realistic AI code generation tasks. DevAI includes rich manual annotations, like a total of 365 hierarchical solution requirements, which make it particularly suitable for an agentic evaluator. We benchmark three of the top code-generating agentic systems using Agent-as-a-Judge and find that our framework dramatically outperforms LLM-as-a-Judge and is as reliable as our human evaluation baseline. Altogether, we believe that this work represents a concrete step towards enabling vastly more sophisticated agentic systems. To help that, our dataset and the full implementation of Agent-as-a-Judge will be publically available at https://github.com/metauto-ai/agent-as-a-judge

Data contamination—the accidental consumption of evaluation examples within the pre-training data—can undermine the validity of evaluation benchmarks. In this paper, we present a rigorous analysis of the effects of contamination on language models at 1B and 8B scales on the machine translation task. Starting from a carefully decontaminated train-test split, we systematically introduce contamination at various stages, scales, and data formats to isolate its effect and measure its impact on performance metrics. Our experiments reveal that contamination with both source and target substantially inflates BLEU scores, and this inflation is 2.5 times larger (up to 30 BLEU points) for 8B compared to 1B models. In contrast, source-only and target-only contamination generally produce smaller, less consistent over-estimations. Finally, we study how the temporal distribution and frequency of contaminated samples influence performance over-estimation across languages with varying degrees of data resources.