A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Hanzhuo Tan, Qi Luo, Jing Li, et al.
LLM4Decompile is the first open-source LLM series trained to decompile binary code, outperforming GPT-4o and Ghidra by over 100% in re-executability.
Zhengren Wang, Jiayang Yu, Dongsheng Ma, et al.
RARE decouples knowledge storage from reasoning by externalizing domain knowledge to retrievable sources and internalizing reasoning patterns, enabling lightweight models to surpass GPT-4 and DeepSeek-R1 by ~20% accuracy.
Yan Hu, Qingyu Chen, Jingcheng Du, et al.
This paper shows that task-specific prompt engineering, incorporating medical knowledge and few-shot examples, significantly improves GPT-3.5 and GPT-4 performance on clinical NER tasks, though still below BioClinicalBERT.
Jürgen Rudolph, Shannon Tan, Samson Tan
This paper compares major chatbots (ChatGPT, Bing Chat, Bard, Ernie) for higher education, finding no A-grade performers despite hype, with GPT-4 leading but Bing Chat and Bard failing.
Unknown
OmniParser converts UI screenshots into structured elements, boosting GPT-4V's ability to interact accurately with interfaces.
Unknown
CriticGPT uses RLHF to train a GPT-4-based model that critiques ChatGPT code outputs, helping humans catch bugs more accurately.
Unknown
NuExtract is a family of small language models fine-tuned on synthetic data to perform structured text-to-JSON extraction, rivaling GPT-4 in accuracy while being orders of magnitude smaller.
Unknown
OmniMath introduces a comprehensive Olympiad-level math benchmark with 4428 problems across 33 sub-domains and 10 difficulty levels, using GPT-4o and an open-source verifier OmniJudge for rigorous evaluation.
Seungone Kim, Juyoung Suk, Shayne Longpre, et al.
Introduces Prometheus, a 13B fully open-source evaluation LLM trained on GPT-4-curated feedback data.
Unknown
GPT-4o is an omni model that accepts and generates text, audio, images, and video.
Unknown
GPT-4V is a multimodal model that integrates text and vision capabilities for analyzing image inputs.
Unknown
LLaVA connects CLIP and Vicuna via end-to-end training on GPT-4-generated instruction-following data from image-text pairs.