ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.
hub
Automatic
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
representative citing papers
TSCG compiles JSON tool schemas into token-efficient structured text, raising tool-use accuracy for small LLMs from 0% to 84.4% on benchmarks while cutting tokens by 52-57%.
CRAFT is a Pareto-front prompt optimizer that allocates scarce LLM validation calls to candidates near the current front using accuracy- and cost-oriented generators plus NSGA-II retention.
GrowLoop proposes a human-seeded self-evolving framework that co-evolves rubrics and cases to evaluate conversational human-likeness with differentiated agreement rules.
iPOE generates and optimizes annotation guidelines from explanations to produce interpretable prompts, reporting up to 39% gains over baselines on four datasets with LLM explanations substituting for human ones.
PQR framework generates diverse realistic queries to elicit QA agent failures, uncovering 23-78% more unhelpful responses than prior methods in e-commerce agent tests.
BrainROI achieves leading cross-subject brain-captioning results on NSD by combining multi-atlas soft-ROI fusion with interpretable prompt optimization.
A single LLM rewrite of skill descriptions using false positive and negative cases matches manual optimization performance in production, with most other pipeline components adding little value.
GBC treats multi-agent LLM workflows as differentiable graphs to enable token-level attribution and targeted optimization, with reported gains on MultiWOZ and τ-bench.
NOVA introduces a level-aware agent harness with architecture gradient and verification cascade to automate recommender architecture evolution while reducing silent failures and human effort.
Empirical study demonstrates that cost-aware skill rewriting for LLM agents can achieve 7% total cost reduction and 6% agent-token cost reduction with preserved quality on SkillsBench.
JTPRO co-optimizes prompts and tool descriptions via reflection to raise overall success rate by 5-20% over baselines on multi-tool benchmarks.
An LM-guided counterfactual pipeline recommends minimal ordinal changes to communication features like tone and actionability, yielding a mean +6.41% gain in predicted positive feedback under independent auditor models.
citing papers explorer
-
ORPO: Monolithic Preference Optimization without Reference Model
ORPO performs preference alignment during supervised fine-tuning via a monolithic odds ratio penalty, allowing 7B models to outperform larger state-of-the-art models on alignment benchmarks.