PEQA method enables sub-4-bit LLM fine-tuning by updating quantization scales, significantly reducing memory demands.

Jeonghoon Kim, Jung Hyun Lee, Sungdong Kim et al.

Proposes LLM-MCTS, combining language model-based world knowledge with search, achieving 91.4% success in household tasks.

Zirui Zhao, Wee Sun Lee, David Hsu

Proposes Temporal Contrastive Learning (TCL) and Siamese STCL to boost low-latency SNNs, achieving SOTA on CIFAR-10/100.

Haonan Qiu, Zeyin Song, Yanqi Chen et al.

Proposes synthetic feedback-based alignment, training ALMoST (65.2% accuracy) without extensive human annotations or proprietary models.

Sungdong Kim, Sanghwan Bae, Jamin Shin et al.

LogicLLM enhances logical reasoning in LLMs via self-supervised training, excelling on ReClor and LogiQA-v2 benchmarks.

Fangkai Jiao, Zhiyang Teng, Bosheng Ding et al.

Proposes GDP-Zero, an open-loop Monte Carlo Tree Search method leveraging prompting large language models for goal-oriented dialogue planning without training.

Xiao Yu, Maximillian Chen, Zhou Yu

Introduces DDPO, a reinforcement learning framework that directly optimizes diffusion models for target-specific objectives, improving image quality and alignment.

Kevin Black, Michael Janner, Yilun Du et al.

Introduces GQA method to convert multi-head models to multi-query models using 5% compute, achieving MQA-like speed and MHA-like quality.

Joshua Ainslie, James Lee-Thorp, Michiel de Jong et al.

Introduces MULTILINGUALHOLISTICBIAS, a dataset extending HOLISTICBIAS to 50 languages, revealing gender translation biases with an average 8 spBLEU gap favoring masculine references.

Marta R. Costa-jussà, Pierre Andrews, Eric Smith et al.

Proposes PMI-based importance masking in bottleneck MAE, boosting dense retrieval performance with 41.7% MRR@10 on MS MARCO.

Zehan Li, Yanzhao Zhang, Dingkun Long et al.

Proposes iEvaLM, an LLM-based interactive evaluation framework, boosting CRS performance and explainability with simulated multi-turn user interactions.

Xiaolei Wang, Xinyu Tang, Wayne Xin Zhao et al.

This study evaluates ChatGPT and GPT-4 as automatic evaluators for abstractive summarization, revealing significant stability and reliability limitations across dimensions and systems.

Chenhui Shen, Liying Cheng, Xuan-Phi Nguyen et al.

ControlVideo employs full cross-frame attention and hierarchical sampling to enable training-free, high-quality controllable text-to-video generation, outperforming state-of-the-art methods.

Yabo Zhang, Yuxiang Wei, Dongsheng Jiang et al.

Developed a sparse representer theorem in RKBS using data-dependent extreme points, promoting solution sparsity.

Rui Wang, Yuesheng Xu, Mingsong Yan

Proposes ALTI-Logit, a residual and attention-based explanation method outperforming gradients and perturbations, with strong results on linguistic and translation tasks.

Javier Ferrando, Gerard I. Gállego, Ioannis Tsiamas et al.

Reinforcement learning with latent motion spaces enables diverse, realistic human motion synthesis in complex 3D indoor scenes.

Kaifeng Zhao, Yan Zhang, Shaofei Wang et al.

Adaptive-Consistency method reduces sampling budget while maintaining high accuracy, with an average accuracy drop of less than 0.1%.

Pranjal Aggarwal, Aman Madaan, Yiming Yang et al.

Pengi unifies audio tasks as text generation, reaching 0.4667 SPIDEr on AudioCaps and 0.9195 accuracy on ESC50.

Soham Deshmukh, Benjamin Elizalde, Rita Singh et al.

CRITIC framework integrates external tools for self-correction, boosting QA, math reasoning, and toxicity reduction by 7-16% improvements.

Zhibin Gou, Zhihong Shao, Yeyun Gong et al.

LLM-Pruner: a structural pruning method that compresses large models with only 50K data in 3 hours, maintaining multi-task performance.

Xinyin Ma, Gongfan Fang, Xinchao Wang