4arXiv cs.CL (Computation and Language)·
JAM: LLM-as-a-Judge framework for theory-agnostic personality recognition via prototypical networks
Researchers introduce JAM (Judge for Adaptive Metric-Alignment), a framework for personality recognition that avoids dependence on predefined psychological taxonomies like Big Five or MBTI. The system uses an Attention-Pooled Graph Prototypical Network with a Cross-Theory Harmonization approach to unify heterogeneous personality datasets, and incorporates an LLM-as-a-Judge mechanism in two configurations to identify ambiguous training samples and guide metric learning. Experiments show improved cross-framework generalization, with the approach enabling inference of latent psychological profiles from text without theory-specific labels.
Related guides (1)
Related events (8)
Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling
This paper identifies and analyzes 'Perceptual Judgment Bias' in multimodal LLM judges, where models anchor on response text rather than visual evidence when the two conflict. The authors introduce a Perceptually Perturbed Judgment Dataset using counterfactual responses to isolate perceptual errors, and a training framework combining GRPO-based reward modeling with batch-ranking objectives. Experiments on MLLM-as-a-Judge benchmarks show improved perceptual fidelity, ranking coherence, and alignment with human evaluation.
Judge Arena: Benchmarking LLMs as Evaluators
Hugging Face and Atla have launched Judge Arena, a platform for benchmarking large language models in their role as automated evaluators. The initiative uses an Elo-based ranking system to compare how well different LLMs judge the quality of model outputs, addressing the growing reliance on LLM-as-judge paradigms in evaluation pipelines. This fills a meta-evaluation gap: as LLM judges become standard practice, understanding their relative reliability and biases becomes critical infrastructure for the field.
Judge-Aware Gated Multi-Task Learning achieves state-of-the-art on UK Employment Tribunal outcome prediction
Researchers propose a Judge-Aware Gated Multi-Task Learning architecture for legal outcome prediction that explicitly disentangles factual case merits from judicial discretion via a gated fusion mechanism conditioned on judge identity. Evaluated on 13,937 UK Employment Tribunal decisions, the approach outperforms supervised fine-tuning of a Gemma-4 26B backbone while requiring an order of magnitude fewer trainable parameters. The key finding is that differentiable structured composition of identity signals outperforms prompt-based composition over a much larger generative model, suggesting conditioning interface choice dominates scale for identity-conditioned classification tasks.
Auditing LLM-as-Judge reliability: judge upgrades are not interchangeable across model families
A new arXiv paper investigates measurement validity problems in LLM-as-judge evaluation, finding that swapping evaluator models changes scores even when candidate responses are fixed. Across four judgment datasets, the authors compare Qwen3 dense judges (1.7B–32B) and MiniMax M2/M2.7 API releases, finding that only the Qwen3 1.7B→4B upgrade yields robust adjacent gains while MiniMax adjacent releases do not. Stronger judges reduce but do not eliminate position and verbosity bias, and repeated-sample juries add little when errors are correlated. The paper argues for standardized reporting requirements including dataset slices, bias probes, error-dependence estimates, and protocol audit trails.
LLM psychological profiles are largely measurement artifacts, not model properties
A new arXiv preprint administers a battery of personality and risk-preference instruments to 56 instruction-tuned LLMs alongside large human reference samples, finding that 81-90% of between-model variation is explained by directional response bias rather than the traits the instruments target. The authors introduce the concept of 'response orthogonality' to explain why some instruments appear more reliable than others, and show that apparent psychological profiles can be manufactured through item selection. The findings challenge the validity of using human-designed psychometric tools to characterize LLMs, with direct implications for safety assessment and the use of LLMs as proxies for human participants in research.
BabelJudge: Benchmark for measuring LLM-as-a-judge reliability across languages and agent trajectories
BabelJudge is a new open-source benchmark and audit framework that systematically measures four failure modes in LLM-as-a-judge systems: position bias, verbosity bias, order inconsistency, and cross-lingual degradation. The framework uses a 'gold-labelling by degradation' technique to generate labeled evaluation pairs without human annotation. Evaluation of Qwen2.5-7B-Instruct-4bit across English, Hindi, Arabic, and Swahili reveals severe cross-lingual reliability drops, with Swahili order consistency collapsing to near-random (0.480). The framework is extended to agentic evaluation with nine trajectory-level perturbations and three new metrics, released as a Python package supporting 11 judge backends.
DEFINED: Data-efficient framework for fine-grained creativity assessment in debate using LLMs
DEFINED is a computational framework for automated creativity assessment in debate scenarios, operationalizing creativity through an eight-dimensional hierarchical metric system implemented via a pretrained autoregressive language model with a hierarchical scoring head. The system addresses data scarcity through constrained data augmentation and mixed-granularity training from limited expert-annotated data. It outperforms prompt-based LLM evaluators and existing debate scoring methods on authentic competition data. The work is relevant to AI evaluation methodology and the broader question of whether LLMs can reliably assess complex human cognitive outputs.
Mistral AI: Using LLM-as-a-Judge with Structured Outputs for RAG Evaluation
Mistral AI published a technical guide on evaluating Retrieval-Augmented Generation (RAG) systems using the 'LLM as a Judge' paradigm combined with their structured outputs API feature. The approach implements the RAG Triad framework—context relevance, groundedness, and answer relevance—using Pydantic schemas to enforce machine-readable evaluation outputs. Mistral models serve as both the generator and judge components, enabling scalable automated evaluation without human annotators.