Natural Language Processing

LLMs: 03-mini achieves superior accuracy without longer reasoning chains, suggesting ’thinking harder’ matters more than ’thinking longer'.

LightThinker: LLMs dynamically compress intermediate steps, reducing memory & boosting reasoning efficiency without sacrificing accuracy.

LLMs struggle with positional bias and lack transparency when summarizing long contexts. This paper introduces SUnsET dataset and fine-tuning methods to improve unstructured evidence citation and summ…

SURVEYX automates academic survey generation, enhancing content and citation quality.

Current LLM evaluation benchmarks often overlook the structural dependencies in multi-turn dialogues, treating them as simple concatenations of single-turn interactions. This approach neglects user in…

Packing new knowledge into LoRA adapters can harm LLMs! A delicate balance is needed to prevent performance decline.

FR-Spec: Frequency-Ranked Speculative Sampling accelerates LLMs by optimizing vocabulary space compression, achieving 1.12x speedup over EAGLE-2.

LLMs have ‘Temporal Heads’ that process time-specific facts!

LORAM: Train small, infer large LLMs by memory-efficient LoRA training. Enables 70B parameter model training on a 20G HBM GPU, replacing A100-80G. Reduces parameter storage cost by 15.81x.

SIFT: Grounds LLM reasoning with ‘Stickers’ to highlight context and improve accuracy without extra training.

REFIND: Detects LLM hallucinations by directly leveraging retrieved documents, using a novel Context Sensitivity Ratio.

MoM: Enhancing linear sequence modeling via mixture-of-memories for improved recall and reduced memory interference.

LongPO: Self-evolve LLMs to excel in long contexts via short-to-long preference optimization, boosting performance without sacrificing short-context skills.

Test-time scaling + confidence = better QA!

CRAW4LLM: Efficiently crawls web pages for LLM pretraining by prioritizing influence scores, boosting data quality & cutting crawling waste.

Autellix: Efficient LLM Serving for Agents

ThinkJSON presents a reinforcement learning strategy to enforce strict schema adherence in LLM generation.

SafeRoute efficiently enhances LLM safety by adaptively using smaller and larger safety guard models, maximizing accuracy while minimizing costs.

Decomposed Reward Models (DRMs) extract diverse human preferences from binary comparisons using PCA, enabling flexible and interpretable LLM alignment.

Perovskite-LLM: a new knowledge-enhanced system boosts perovskite solar cell research by integrating a domain-specific knowledge graph, high-quality datasets, and specialized LLMs for superior knowled…