Weaver builds conditional proposal trees from a factorized drafter’s top-K marginals and a rollback-free GDN tree-verify kernel, yielding 4.37× speedup over AR decoding.
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL) , year=
9 Pith papers cite this work, alongside 40 external citations. Polarity classification is still indexing.
9
Pith papers citing it
40
external citations · OpenAlex
citation-role summary
background 1
citation-polarity summary
years
2026 9roles
background 1polarities
unclear 1representative citing papers
SCICONVBENCH is a new benchmark evaluating LLMs on multi-turn disambiguation and inconsistency resolution for task formulation in computational science, with frontier models reaching only 52.7% success on fluid mechanics disambiguation cases.
R²A uses a hybrid ensemble surrogate router and suffix optimization to significantly increase black-box LLM router selection of expensive models across query distributions.
FrontierFinance benchmark shows human financial experts outperform state-of-the-art LLMs by achieving higher scores and more client-ready outputs on realistic long-horizon tasks.
IHDec applies JSD-steered contrastive decoding to enforce multi-turn instruction hierarchies in LLMs without fine-tuning.
COALA applies convex optimization reformulations of neural networks to direct preference optimization, claiming single-GPU training with ~18% of DPO's TFLOPs and competitive performance on multiple datasets and models up to 8B parameters.
SafeMoE isolates unsafe knowledge in domain-specific LoRA experts and routes them via a lightweight gate trained on safe responses to produce safer and more informative LLM outputs with zero-shot generalization.
LoCar is a localization-aware evaluation framework for in-vehicle assistants that identifies unstable Korean honorific control and weaker performance on strategic metrics like clarification and proactivity in current LLMs.
Interactive LLM dialogue raised residents' hard-case diagnostic correctness from 0.589 to 0.734 and produced medium effect sizes in a blinded study of seven physicians on 52 emergency cases.
citing papers explorer
-
Trees from Marginals: Autoregressive drafting with factorized priors
Weaver builds conditional proposal trees from a factorized drafter’s top-K marginals and a rollback-free GDN tree-verify kernel, yielding 4.37× speedup over AR decoding.
-
SCICONVBENCH: Benchmarking LLMs on Multi-Turn Clarification for Task Formulation in Computational Science
SCICONVBENCH is a new benchmark evaluating LLMs on multi-turn disambiguation and inconsistency resolution for task formulation in computational science, with frontier models reaching only 52.7% success on fluid mechanics disambiguation cases.
-
Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization
R²A uses a hybrid ensemble surrogate router and suffix optimization to significantly increase black-box LLM router selection of expensive models across query distributions.
-
FrontierFinance: A Long-Horizon Computer-Use Benchmark of Real-World Financial Tasks
FrontierFinance benchmark shows human financial experts outperform state-of-the-art LLMs by achieving higher scores and more client-ready outputs on realistic long-horizon tasks.
-
IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchies
IHDec applies JSD-steered contrastive decoding to enforce multi-turn instruction hierarchies in LLMs without fine-tuning.
-
Convex Optimization for Alignment and Preference Learning on a Single GPU
COALA applies convex optimization reformulations of neural networks to direct preference optimization, claiming single-GPU training with ~18% of DPO's TFLOPs and competitive performance on multiple datasets and models up to 8B parameters.
-
Dialectics of Alignment: Harnessing Unsafe Knowledge for Dynamic Safety Routing
SafeMoE isolates unsafe knowledge in domain-specific LoRA experts and routes them via a lightweight gate trained on safe responses to produce safer and more informative LLM outputs with zero-shot generalization.
-
LoCar: Localization-Aware Evaluation of In-Vehicle Assistants through Fine-Grained Sociolinguistic Control
LoCar is a localization-aware evaluation framework for in-vehicle assistants that identifies unstable Korean honorific control and weaker performance on strategic metrics like clarification and proactivity in current LLMs.
-
Human-LLM Dialogue Improves Diagnostic Accuracy in Emergency Care
Interactive LLM dialogue raised residents' hard-case diagnostic correctness from 0.589 to 0.734 and produced medium effect sizes in a blinded study of seven physicians on 52 emergency cases.