Hacker News with Generative AI: AI Research

An OpenAI researcher who worked on GPT-4.5 had their green card denied (techcrunch.com)
Kai Chen, a Canadian AI researcher working at OpenAI who’s lived in the U.S. for 12 years, was denied a green card, according to Noam Brown, a leading research scientist at the company.
Immigration, AI Research, OpenAI, Green Card
11 points by ydnyshhh 10 hours ago | 1 comments
The State of Reinforcement Learning for LLM Reasoning (sebastianraschka.com)
A lot has happened this month, especially with the releases of new flagship models like GPT-4.5 and Llama 4. But you might have noticed that reactions to these releases were relatively muted. Why? One reason could be that GPT-4.5 and Llama 4 remain conventional models, which means they were trained without explicit reinforcement learning for reasoning.
Reinforcement Learning, Generative AI, AI Research
9 points by jonbaer 5 days ago | 0 comments
Inferring the Phylogeny of Large Language Models (arxiv.org)
This paper introduces PhyloLM, a method adapting phylogenetic algorithms to Large Language Models (LLMs) to explore whether and how they relate to each other and to predict their performance characteristics.
Machine Learning, Generative AI, AI Research
68 points by weinzierl 7 days ago | 6 comments
NoProp: Training neural networks without back-propagation or forward-propagation (arxiv.org)
The canonical deep learning approach for learning requires computing a gradient term at each layer by back-propagating the error signal from the output towards each learnable parameter.
Neural Networks, Machine Learning, Deep Learning, AI Research, Computer Science
161 points by belleville 12 days ago | 49 comments
Anthropic: Circuit Tracing and on the Biology of a Large Language Model [video] (youtube.com)
Generative AI, AI Research, Video, Anthropic
6 points by swyx 16 days ago | 0 comments
Obituary for Cyc (yuxi-liu-wired.github.io)
The legendary Cyc project, Douglas Lenat’s 40-year quest to build artificial general intelligence by scaling symbolic logic, has failed.
Artificial Intelligence, History, Symbolic Logic, AI Research
449 points by todsacerdoti 17 days ago | 282 comments
Meta's surprise Llama 4 drop exposes the gap between AI ambition and reality (arstechnica.com)
On Saturday, Meta released its newest Llama 4 multimodal AI models in a surprise weekend move that caught some AI experts off guard.
Generative AI, Artificial Intelligence, Meta, AI Research
9 points by pera 18 days ago | 0 comments
UCSD: Large Language Models Pass the Turing Test (arxiv.org)
We evaluated 4 systems (ELIZA, GPT-4o, LLaMa-3.1-405B, and GPT-4.5) in two randomised, controlled, and pre-registered Turing tests on independent populations.
Generative AI, Turing Test, AI Research
91 points by Mossy9 24 days ago | 106 comments
Attention is NOT all you need: Qwerky-72B trained using only 8 AMD MI300X GPUs (recursal.ai)
We are proud to announce the updated Qwerky-72B and 32B.
Machine Learning, Hardware, AI Research
20 points by jtatarchuk 24 days ago | 3 comments
Tied Crosscoders: Tracing How Chat LLM Behavior Emerges from Base Model (lesswrong.com)
We are interested in model-diffing: finding what is new in the chat model when compared to the base model. One way of doing this is training a crosscoder, which would just mean training an SAE on the concatenation of the activations in a given layer of the base and chat model. When training this crosscoder, we find some latents whose decoder vector mostly helps reconstruct the base model activation and does not affect the reconstruction for the chat model activation.
Generative AI, Machine Learning, AI Research, Neural Networks
7 points by aranguri 33 days ago | 0 comments
Beyond Diffusion: Inductive Moment Matching (lumalabs.ai)
There is a growing sentiment in the AI community that generative pre-training is reaching a limit. However, we argue that these limits are not due to a lack of data itself, but rather a stagnation in algorithmic innovation.
Generative AI, Machine Learning, AI Research
202 points by outrun86 45 days ago | 31 comments
GPT-4 Admit to Progressive Restrictions – OpenAI Subsidies Degrading Performance (reddit.com)
OpenAI is an AI research and deployment company. OpenAI's mission is to ensure that artificial general intelligence benefits all of humanity. We are an unofficial community. OpenAI makes ChatGPT, GPT-4, and DALL·E 3.
Artificial Intelligence, OpenAI, AI Research
4 points by PrixKc 51 days ago | 1 comments
OpenAI expands Deep Research to all paying ChatGPT users (engadget.com)
When OpenAI announced Deep Research at start of February, the company promised to bring the tool to Plus users "in about a month," and now it's doing exactly that.
Generative AI, OpenAI, ChatGPT, AI Research, Software
22 points by thundergolfer 59 days ago | 5 comments
Stone Soup AI (2024) (simons.berkeley.edu)
For some time, I’ve argued that a common conception of AI is misguided.
Artificial Intelligence, AI Research, Opinion
145 points by dredmorbius 60 days ago | 65 comments
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 81 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 86 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 87 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 90 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 97 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 119 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 133 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 135 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 136 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 165 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 166 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 172 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 267 days ago | 11 comments
Large Enough (mistral.ai)
Generative AI, AI Research
639 points by davidbarker 276 days ago | 496 comments
Chat with Meta Llama 3.1 405B (replicate.dev)
Chatbots, Generative AI, AI Research, Open Source
13 points by bfirsh 276 days ago | 1 comments
ChatGPT is better at generating code for problems written before 2021 (ieee.org)
Generative AI, Programming, AI Research
59 points by ummonk 292 days ago | 46 comments