Ruchir Narfk @narfkin

In a recent technical report, LearnLM, our set of AI models and capabilities fine-tuned for learning, outperformed other leading AI models on the principles of learning science. Now it’s available to try out in AI Studio. Learn more ↓ goo.gle/4gmEdxp

Introducing ASAL: Automating the Search for Artificial Life with Foundation Models sakana.ai/asal/ Artificial Life (ALife) research holds key insights that can transform and accelerate progress in AI. By speeding up ALife discovery with AI, we accelerate our understanding of emergence, evolution, and intelligence–core principles that can inspire the next generation of AI systems! We proudly collaborated with MIT, OpenAI, Swiss AI Lab IDSIA, and Ken Stanley on this exciting project. Full Paper (Website): pub.sakana.ai/asal/ Full Paper (arxiv): asal.sakana.ai/paper/ Code: github.com/SakanaAI/asal/ In this work, we propose a new algorithm called Automated Search for Artificial Life (“ASAL”) to automate the discovery of artificial life using vision-language foundation models. Instead of tediously hand-designing every tiny rule of an Alife simulation, simply describe the space of simulations to search over, and ASAL will automatically discover the most interesting and open-ended artificial lifeforms! Because of the generality of foundation models, ASAL can discover new lifeforms across a diverse range of seminal ALife simulations, including Boids, Particle Life, Game of Life, Lenia, and Neural Cellular Automata. ASAL even discovered novel cellular automata rules that are more open-ended and expressive than the original Conway’s Game of Life. We believe this new paradigm may reignite ALife research by overcoming the bottleneck of manually designed simulations, thus advancing beyond the limits of human ingenuity.

Excited to see OpenAI's recent project, MLE-bench, is based on our open-source effort, AIDE. In their independent evaluation, AIDE surpasses other MLE agents by a large margin!

OpenAI @OpenAI

2 years ago

We’re releasing a new benchmark, MLE-bench, to measure how well AI agents perform at machine learning engineering. The benchmark consists of 75 machine learning engineering-related competitions sourced from Kaggle. openai.com/index/mle-benc…

208 516 4K 845K 858

What will a world look like with 100 billion digital human beings? Today we share our tech report on Project Sid – a glimpse at the first AI agent civilization (powered by our new PIANO architecture). github.com/altera-al/proj… 1/8

As a RL research myself, I once doubted Reinforcement Learning (RL) because massive self-supervised LLMs were dominating. But now I see how RL can bring us closer to super-intelligent (ASI) systems—far beyond board games. Here’s what changed my mind: (1/5)

x.com/i/article/1874…

Happy 1^3+2^3+3^3+4^3+5^3+6^3+7^3+8^3+9^3 (2025)

Since talking to @JAldrichPL on here is the only way I know of to get useful communication with the ACM pubs board, I will just mention that this is very disappointing from the ACM OPEN perspective: github.com/borisveytsman/…

Inument hosted an engaging in-person tech session on “Crafting the Ideal Software Architecture: Lessons from Complex Systems” at our Dhaka office, led by Mahmudul Islam Romel, CloudFinOps Tech Lead at Rokt.

keyboard_arrow_left Previous keyboard_arrow_right Next

AI is no longer a thing of the future but a present-day reality that has found its way into most industries across the world. inument.com/how-ai-is-chan…

### AI for Leaders at the National University of Singapore ### Our CEO recently completed an insightful week at NUS’s “AI for Leaders: Leveraging Data Analytics for Business” program. This explored embedding AI into business, ML, & generative AI for effective decision-making.

keyboard_arrow_left Previous keyboard_arrow_right Next

Releasing AutoCodeRover-v2 fully open source available for download by anyone ! autocoderover.dev Achieving 24.9% on SWE-bench full and 46.2% on SWE-bench Verified! **Source code available** at github.com/nus-apr/auto-c…! We are releasing the results here, while still waiting for SWE-bench leaderboard to process our submission. Timeline: - Nov 8: We opened our [PR #111](github.com/swe-bench/expe…); - Nov 17-18: SWE-bench merged 6 **more recent** PRs (#112-116, #121) - Nov 19: We sent a query to SWE-bench in our PR & over email. - Nov 20: We sent another query in our PR. - Nov 21-23: We have *not* received any response from SWE-bench. Our [last submission](github.com/swe-bench/expe…) also took **45+** days to be processed. Stay tuned by following us on X (@autocoderover), [discord](discord.gg/c6DAfNUngZ), and checking our [website](autocoderover.net)! Reach out to us directly! #SWEbench #AutoCodeRover

NUS retains its place among the world’s 20 best universities in the @timeshighered World University Rankings 2024! Congrats & well done to our community of faculty, staff & students in pursuing future-focused education, research and innovation. #NUSExcellence #ShapeTheFuture

🏆Kudos Prof @AbhikRoychoudh1 & team for winning Most Influential Paper at #ICSE2023. Their work is reshaping code repair & security, and has ignited a 5-year APR program funded by the @MOEsg. Congrats! 🎉Story: comp.nus.edu.sg/news/2023-icse… #NUSExcellence #meetourfaculty

🚀 Concerned about data safety? Prof @trevorcarlson & team at NUS Computing created “AfterImage” to study device security risks. Their work unlocks insights into device vulnerabilities & ways to shield against cyber threats! Story: comp.nus.edu.sg/news/features/… #Meetourfaculty

You can now use @OpenAI’s new o1 model in Copilot Pro, Business, and Enterprise—and in GitHub Models. It’s all about that developer choice. 💻 github.blog/news-insights/…

Splean: a simple Separation Logic in Lean github.com/verse-lab/sple…

AI deepfakes 🤖 are becoming increasingly prevalent. See what #NUSComputing's professors Terence Sim and Hahn Jungpil have to say about this new, worrying trend. ST article📰: bit.ly/3NYSa93 LZHB article📰: bit.ly/3JOpc9w #AI #Deepfakes #meetourfaculty

keyboard_arrow_left Previous keyboard_arrow_right Next

AutoCodeRover: Autonomous SE Do not see a software project merely as files. Do not see software engineering as a bunch of prompts to LLMs. Do software engineering like engineers do, by navigating code base iteratively! And more autonomous !! github.com/nus-apr/auto-c… [3 / 4]