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.
Andrés Marafioti, Orr Zohar, Miquel Farr'e, et al.
SmolVLM introduces a family of compact vision-language models achieving strong performance on resource-constrained devices through aggressive token compression and optimized encoder-LM balance.
Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, et al.
Code Llama is a family of open-source LLMs for code based on Llama 2, achieving state-of-the-art performance in code generation, infilling, and long-context tasks.
Krzysztof Czarnecki, Ulrich W. Eisenecker
This book introduces generative programming, a paradigm for automating software component assembly using domain engineering, feature modeling, and code generation.
Nathan Shone, Trần Nguyên Ngọc, Vu Dinh Phai, et al.
Proposes a nonsymmetric deep autoencoder (NDAE) for unsupervised feature learning and a stacked NDAE classifier for network intrusion detection, evaluated on KDD Cup '99 and NSL-KDD datasets.
Jonghyun Lee, Dae Won Jun, Ildae Song, et al.
DLM-DTI uses a hint-based learning strategy to create a compact and efficient target encoder for drug-target interaction prediction, reducing VRAM usage to 7.7GB.
Riccardo Miotto, Li Li, Brian Kidd, et al.
Unsupervised deep feature learning with stacked denoising autoencoders on EHRs of ~700k patients yields a general-purpose patient representation that significantly improves predictive modeling for 78 diseases.
Iulian Vlad Serban, Alessandro Sordoni, Yoshua Bengio, et al.
Extends hierarchical recurrent encoder-decoder networks to open-domain dialogue, showing competitiveness with n-gram models and improvement via bootstrapping from QA pairs and pretrained embeddings.
Michelle Brachman, Amina El-Ashry, Casey Dugan, et al.
A survey study of knowledge workers reveals current LLM uses like code generation and text improvement, and a future vision of integrated workflow and data interaction.
Yifei Zhou, Sergey Levine, J. Weston, et al.
Self-Challenging framework enables LLM agents to self-generate high-quality training tasks via Code-as-Task, achieving over two-fold improvement on tool-use benchmarks without human annotation.
Unknown
CriticGPT uses RLHF to train a GPT-4-based model that critiques ChatGPT code outputs, helping humans catch bugs more accurately.
Unknown
A third-party, objective comparison of OpenAI GPT and Google Gemini models with reproducible code and transparent results.