PROMETHEUS: Fine-Grained Evaluation Capability in Language Models
PROMETHEUS introduces fine-grained evaluation capabilities in language models, emphasizing the importance of open-source and reproducible models.
AnyGPT: Unified Multimodal Language Model with Discrete Sequence Modeling
AnyGPT introduces a unified multimodal language model using discrete representations for processing various modalities seamlessly.
Contrastive Activation Addition for Steering Language Models
Introducing CAA for precise steering of language models by modifying activations, enhancing alignment techniques.
LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon Gameplay
LLM-based agents exhibit adaptive social behaviors in Avalon gameplay, showcasing potential in dynamic interactions.
PALO: A Polyglot Large Multimodal Model for 5B People
PALO introduces a Large Multilingual Multimodal Model to bridge the gap in vision-language tasks across ten major languages, emphasizing inclusivity and performance improvements.
WMDP Benchmark: Unlearning for Hazardous Knowledge
공개적인 위험 지식을 줄이기 위한 WMDP 벤치마크와 CUT 방법 소개
Design2Code: Automating Front-End Engineering Benchmark Study
Generative AI advancements enable automating front-end engineering with Design2Code task.
LLM Safety Enhancement through Constrained Direct Preference Optimization
Constrained DPO (C-DPO) enhances LLM safety efficiently and effectively.
HeteGen: Heterogeneous Parallel Inference for Large Language Models on Resource-Constrained Devices
HeteGen introduces a novel approach for efficient inference on resource-constrained devices by leveraging heterogeneous parallel computing, demonstrating significant improvements in speed.
LARA: Linguistic-Adaptive Retrieval-Augmented LLMs for Multi-Turn Intent Classification
Enhancing multi-turn intent classification with Linguistic-Adaptive Retrieval-Augmented Language Models (LLMs).
© 2024 by Linnk AI