Hacker News with Generative AI: AI Research

DeepRAG: Thinking to retrieval step by step for large language models (arxiv.org)
Large Language Models (LLMs) have shown remarkable potential in reasoning while they still suffer from severe factual hallucinations due to timeliness, accuracy, and coverage of parametric knowledge.

Generative AI, AI Research

191 points by fofoz 18 days ago | 29 comments

Scaling the Tülu 3 post-training recipes to surpass the perf of DeepSeek V3 (allenai.org)
Following the success of our Tülu 3 release in November, we are thrilled to announce the launch of Tülu 3 405B—The first application of fully open post-training recipes to the largest open-weight models. With this release, we demonstrate the scalability and effectiveness of our post-training recipe applied at 405B parameter scale.

Generative AI, AI Research, Open Source

14 points by Philpax 23 days ago | 0 comments

OpenAI: DeepSeek "found some of the core ideas that we did on our way to o1" (twitter.com)

OpenAI, AI Research

9 points by modeless 25 days ago | 6 comments

Bespoke-Stratos: The unreasonable effectiveness of reasoning distillation (bespokelabs.ai)
We trained Bespoke-Stratos-32B, our reasoning model distilled from DeepSeek-R1 using Berkeley NovaSky’s Sky-T1 data pipeline.

Generative AI, Machine Learning, Reasoning, AI Research

13 points by madiator 27 days ago | 2 comments

FrontierMath Was Funded by OpenAI (lesswrong.com)
FrontierMath was funded by OpenAI.

FrontierMath, OpenAI, AI Research, Funding

22 points by s-macke 34 days ago | 4 comments

Explaining Large Language Models Decisions Using Shapley Values (arxiv.org)
The emergence of large language models (LLMs) has opened up exciting possibilities for simulating human behavior and cognitive processes, with potential applications in various domains, including marketing research and consumer behavior analysis.

Generative AI, Explainability, AI Research, Machine Learning

89 points by veryluckyxyz 56 days ago | 19 comments

Ilya Sutskever NeurIPS talk [video] (youtube.com)
OpenAI’s cofounder and former chief scientist, Ilya Sutskever, made headlines earlier this year after he left to start his own AI lab called Safe Superintelligence Inc.

Artificial Intelligence, AI Research, Conferences, Machine Learning

309 points by mfiguiere 70 days ago | 240 comments

Ethical Challenges Related to the NeurIPS 2024 Best Paper Award (var-integrity-report.github.io)
To AI Research Community: This report is written to convey our serious concerns about the recent recipient of the Best Paper award at NeurIPS 2024, Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction (VAR) . While we acknowledge that this NeurIPS paper is technically sound, we must emphasize that it involves serious misconduct by the first author (Keyu Tian), which fundamentally undermines the core values of integrity and trust upon which our academic community is built.

AI Research, Ethics, Academic Integrity, Scientific Misconduct, NeurIPS

30 points by Dowwie 72 days ago | 12 comments

Ethical Challenges Related to the NeurIPS 2024 Best Paper Award (var-integrity-report.github.io)
To AI Research Community: This report is written to convey our serious concerns about the recent recipient of the Best Paper award at NeurIPS 2024, Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction (VAR). While we acknowledge that this NeurIPS paper is technically sound, we must emphasize that it involves serious misconduct by the first author (Keyu Tian), which fundamentally undermines the core values of integrity and trust upon which our academic community is built.

AI Research, Ethics, Academic Integrity

13 points by baobabKoodaa 73 days ago | 1 comments

The Lost Reading Items of Ilya Sutskever's AI Reading List (tensorlabbet.com)
In this post: An attempt to reconstruct Ilya Sutskever's 2020 AI reading list (8 min read)

Artificial Intelligence, Reading Lists, Research, AI Research, Computer Science

45 points by tarolangner 102 days ago | 6 comments

GPTs Are Maxed Out (thealgorithmicbridge.com)
March 2024. OpenAI CEO Sam Altman joins podcaster Lex Fridman for the second time since ChatGPT came out a year prior. The stakes are high and anticipation is tangible. GPT-5 appears to be around the corner. Altman, elusive as always, provides only one data point for us hungry spectators: The next-gen model (he doesn’t name it) will be better than GPT-4 to the same degree that GPT-4 was better than GPT-3.

Generative AI, Artificial Intelligence, AI Research, Technology

32 points by pseudolus 103 days ago | 81 comments

Hunyuan-Large: An Open-Source Moe Model with 52B Activated Parameters (arxiv.org)
In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens.

Generative AI, Open Source, AI Research

5 points by belter 109 days ago | 0 comments

Meta FAIR refuses to cite a pre-existing open-source project – to claim novelty (granadacoders.es)

Open Source, AI Research, Ethics, Meta

20 points by keskival 204 days ago | 11 comments

Large Enough (mistral.ai)

Generative AI, AI Research

639 points by davidbarker 213 days ago | 496 comments

Chat with Meta Llama 3.1 405B (replicate.dev)

Chatbots, Generative AI, AI Research, Open Source

13 points by bfirsh 214 days ago | 1 comments

ChatGPT is better at generating code for problems written before 2021 (ieee.org)