Derives a blockwise resolvent-style attention operator that exploits structured sparsity for subquadratic O(n^{4/3}d) entity tracking while matching dense accuracy.
End-to-end Neural Coreference Resolution
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
Later LLM layers align better with human cognitive effort in syntactic ambiguity than early layers do, indicating dual processing modes and complementary benefits from multi-layer probability updates.
EMERGE is a benchmark dataset of 233K Wikipedia passages paired with 1.45 million Wikidata edit operations across seven yearly snapshots from 2019 to 2025 for evaluating knowledge graph updates from emerging text.
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
SuperGLUE is a new benchmark with more difficult language understanding tasks, a toolkit, and leaderboard to drive further progress beyond GLUE.
The 2026 multilingual coreference shared task expanded to 27 datasets in 19 languages with a focus on long-range entities; traditional systems led but LLMs showed notable potential.
Two-stage multilingual then dataset-specific adapter fine-tuning of Gemma-3-27b with headword XML mention representation and iterative annotation achieved first place in the CRAC 2026 LLM track.
citing papers explorer
-
Structured-Sparse Attention for Entity Tracking with Subquadratic Sequence Complexity
Derives a blockwise resolvent-style attention operator that exploits structured sparsity for subquadratic O(n^{4/3}d) entity tracking while matching dense accuracy.
-
Dual Alignment Between Language Model Layers and Human Sentence Processing
Later LLM layers align better with human cognitive effort in syntactic ambiguity than early layers do, indicating dual processing modes and complementary benefits from multi-layer probability updates.
-
EMERGE: A Benchmark for Updating Knowledge Graphs with Emerging Textual Knowledge
EMERGE is a benchmark dataset of 233K Wikipedia passages paired with 1.45 million Wikidata edit operations across seven yearly snapshots from 2019 to 2025 for evaluating knowledge graph updates from emerging text.
-
LIMO: Less is More for Reasoning
LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.
-
SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems
SuperGLUE is a new benchmark with more difficult language understanding tasks, a toolkit, and leaderboard to drive further progress beyond GLUE.
-
Findings of the Fifth Shared Task on Multilingual Coreference Resolution: Expanding Datasets for Long-Range Entities
The 2026 multilingual coreference shared task expanded to 27 datasets in 19 languages with a focus on long-range entities; traditional systems led but LLMs showed notable potential.
-
Closing the Gap at CRAC 2026: Two-Stage Adaptation for LLM-Based Multilingual Coreference Resolution
Two-stage multilingual then dataset-specific adapter fine-tuning of Gemma-3-27b with headword XML mention representation and iterative annotation achieved first place in the CRAC 2026 LLM track.