benchmark
Multi-Round Coreference Resolution
benchmarkactiveprovisional
multi-round-coreference-resolution-f1c0305a·1 events·first seen 47h agoAliases: Multi-Round Coreference Resolution
Co-occurring entities
More like this (12)
Multilingual Coreference Resolution Shared TaskCORE (Contrastive Reflection)Reeve Foundation Multilingual CorpusAccuracy and Satisfaction in Multi-Turn LLM Dialogues for NFR AssessmentLong-context Reasoning BenchmarksMulti-hop Question AnsweringContext-Driven Incremental Compression for Multi-Turn Dialogue GenerationContext-Driven Incremental Compression for Multi-Turn Dialogue GenerationMulti-Task Bayesian In-Context LearningWhen the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning Modelsmulti-round event injectionUncertainty-Aware Hybrid Retrieval for Long-Document RAG
Recent events (1)
Randomized YaRN improves LLM length generalization for long-context reasoning
Researchers propose Randomized YaRN, a training method that combines YaRN-based positional extrapolation with randomized positional encodings and a length curriculum to improve LLM generalization to long contexts. Models trained on sequences under 8K tokens show consistent reasoning improvements on context lengths from 16K to 128K on BABILong and MRCR benchmarks. The key insight is that exposing models to out-of-distribution positional representations during short-context training enables better generalization at far longer inference-time lengths.