Related papers: Aligning Language Model Benchmarks with Pairwise P…
Automated negotiation in complex, multi-party and multi-issue settings critically depends on accurate opponent modeling. However, conventional numerical-only approaches fail to capture the qualitative information embedded in natural…
Benchmarking has long served as a foundational practice in machine learning and, increasingly, in modern AI systems such as large language models, where shared tasks, metrics, and leaderboards offer a common basis for measuring progress and…
Post-training alignment optimizes language models to match human preference signals, but this objective is not equivalent to modeling observed human behavior. We compare 120 base-aligned model pairs on more than 10,000 real human decisions…
Aligning large language models (LLMs) to human preferences is a crucial step in building helpful and safe AI tools, which usually involve training on supervised datasets. Popular algorithms such as Direct Preference Optimization (DPO) rely…
Benchmarks have emerged as the central approach for evaluating Large Language Models (LLMs). The research community often relies on a model's average performance across the test prompts of a benchmark to evaluate the model's performance.…
Large language models~(LLMs) are expected to be helpful, harmless, and honest. In different alignment scenarios, such as safety, confidence, and general preference alignment, binary preference data collection and reward modeling are…
Large language model (LLM) evaluation is increasingly costly, prompting interest in methods that speed up evaluation by shrinking benchmark datasets. Benchmark prediction (also called efficient LLM evaluation) aims to select a small subset…
The rapid proliferation of benchmarks for evaluating large language models (LLMs) has created an urgent need for systematic methods to assess benchmark quality itself. We propose Benchmark^2, a comprehensive framework comprising three…
Large language models (LLMs) have demonstrated remarkable capabilities in handling complex dialogue tasks without requiring use case-specific fine-tuning. However, analyzing live dialogues in real-time necessitates low-latency processing…
As language models accelerate scientific research by automating hypothesis generation and implementation, a new bottleneck emerges: evaluating and filtering hundreds of AI-generated ideas without exhaustive experimentation. We ask whether…
Word alignment is essential for the downstream cross-lingual language understanding and generation tasks. Recently, the performance of the neural word alignment models has exceeded that of statistical models. However, they heavily rely on…
Reward models are used throughout the post-training of language models to capture nuanced signals from preference data and provide a training target for optimization across instruction following, reasoning, safety, and more domains. The…
Existing benchmarks for evaluating mathematical reasoning in large language models (LLMs) rely primarily on competition problems, formal proofs, or artificially challenging questions -- failing to capture the nature of mathematics…
Instruction-following is a foundational capability of large language models (LLMs), with its improvement hinging on scalable and accurate feedback from judge models. However, the reliability of current judge models in instruction-following…
Two methodologies dominate current practices of benchmarking: rubric-based scoring evaluates items against predefined criteria, whereas comparative judgment elicits pairwise preferences between outputs. Although both methodologies are…
As large language models become increasingly capable of generating code, evaluating their performance remains a complex and evolving challenge. Existing benchmarks primarily focus on functional correctness, overlooking the diversity of…
Recent advancements in text-to-speech (TTS) have shown that language model (LM)-based systems offer competitive performance to their counterparts. Further optimization can be achieved through preference alignment algorithms, which adjust…
The goal of aligning language models to human preferences requires data that reveal these preferences. Ideally, time and money can be spent carefully collecting and tailoring bespoke preference data to each downstream application. However,…
We propose a novel preference alignment framework for improving spoken dialogue models on real-time conversations from user interactions. Current preference learning methods primarily focus on text-based language models, and are not…
Evaluation of large language models for code has primarily relied on static benchmarks, including HumanEval (Chen et al., 2021), or more recently using human preferences of LLM responses. As LLMs are increasingly used as programmer…