2025-02-19s

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.

RealSyn: A new, scalable multimodal dataset revolutionizes vision-language learning by effectively using interleaved image-text documents.

ARM4R pre-trains autoregressive robotic models using low-level 4D representations from human videos, achieving efficient transfer learning and improved task performance across various environments.

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…

PAFT dynamically adjusts prompts during LLM fine-tuning, improving model robustness and generalization across diverse prompts without sacrificing performance or efficiency.

mmMamba: a novel framework creates linear-complexity multimodal models via distillation, drastically improving efficiency without sacrificing performance.

Magma: a new foundation model for multimodal AI agents excels at bridging verbal and spatial intelligence, achieving state-of-the-art performance across various tasks, including UI navigation and robo…

HEADINFER achieves memory-efficient LLM inference by cleverly offloading key-value cache to the CPU, enabling 4 million token inference on a single consumer GPU.

Eager updates drastically speed up training massive language models by cleverly overlapping communication and computation in DiLoCo, achieving near-optimal performance even with low bandwidth.

Crowd-based comparative evaluation significantly boosts LLM-as-a-judge accuracy by using crowd responses to expose deeper details, resulting in more reliable and efficient auto-evaluation.

LLMs can losslessly compress 1568 tokens into a single vector, surpassing prior methods by two orders of magnitude.

Contrary to popular belief, longer reasoning chains don’t always boost Large Language Model (LLM) accuracy; this research reveals that parallel scaling with shorter solutions outperforms sequential sc…

FLAG-TRADER fuses LLMs & RL for enhanced financial trading, achieving superior performance compared to traditional methods by efficiently integrating multimodal data and adapting to market dynamics.

RDLM: A novel continuous diffusion model for language modeling leverages the geometry of categorical distributions, outperforming existing discrete approaches and approaching autoregressive model perf…

Atom of Thoughts (AOT) revolutionizes LLM test-time scaling by decomposing complex reasoning into independent sub-questions, drastically reducing computation while maintaining high accuracy.

FinMTEB: A new benchmark reveals that general-purpose embedding models struggle in the finance domain; domain-specific models excel, and surprisingly, simple BoW outperforms sophisticated models on ce…

XLM-SWCM: A novel framework efficiently adapts multilingual encoders for text generation in extremely low-resource languages by cleverly sharing weights between encoder and decoder, achieving superior…

This survey paper comprehensively analyzes methods for injecting domain-specific knowledge into LLMs, categorizing them into four key approaches and evaluating their trade-offs to enhance performance …

HealthGPT: A novel medical vision-language model unifying comprehension and generation via heterogeneous knowledge adaptation, achieving superior performance on various medical tasks.