Mononito Goswami
@MononitoGoswami
Coding Agents at Applied Scientist @AmazonScience | PhD @CarnegieMellon | Ex @GoogleAI
🎓 I recently graduated with my Ph.D at @CarnegieMellon, and started as an Applied Scientist @AmazonScience working on coding agents 🧑💻! I'm grateful for everyone who made this journey possible ❤️
Introducing Kiro, an all-new agentic IDE that has a chance to transform how developers build software. Let me highlight three key innovations that make Kiro special: 1 - Kiro introduces spec-driven development, helping developers express their intent clearly through natural…
Releasing mini, a radically simple SWE-agent: 100 lines of code, 0 special tools, and gets 65% on SWE-bench verified! Made for benchmarking, fine-tuning, RL, or just for use from your terminal. It’s open source, simple to hack, and compatible with any LM! Link in 🧵
🔥 Excited to announce Mitra, our new tabular foundation model, featured on Amazon Science! Mitra is pretrained purely on synthetic data and archives SOTA performance on real tabular datasets.
Introducing Mitra: a foundation model from Amazon researchers that outperforms traditional methods for tabular data by learning from diverse synthetic priors. Mitra uses in-context learning to adapt to new tasks without separate models for each dataset: amazon.science/blog/mitra-mix…
New Anthropic Research: “Inverse Scaling in Test-Time Compute” We found cases where longer reasoning leads to lower accuracy. Our findings suggest that naïve scaling of test-time compute may inadvertently reinforce problematic reasoning patterns. 🧵
Introducing Mitra: a foundation model from Amazon researchers that outperforms traditional methods for tabular data by learning from diverse synthetic priors. Mitra uses in-context learning to adapt to new tasks without separate models for each dataset: amazon.science/blog/mitra-mix…
Foundation Models for Structured Data Workshop, Today, ICML 2025, starting at 9 AM PT, West Ballroom D, best room in the building! Let's do this!
🚨 Join us tomorrow at the 1st ICML Workshop on Foundation Models for Structured Data! 📅 July 18 | ⏰ 9 AM - 5 PM 📍 West Ballroom D 🎉 70 accepted papers 🎤 8 spotlight orals 🌟 6 invited talks from top experts in tabular & timeseries data! #ICML2025 #FMSD
At @icmlconf today, @inverse_hessian and I will share our thoughts on what time series foundation models (No, not LLMs) really learn, how to steer them effectively, and making mechanistic interpretability useful 👀 4:30 PM, West Exhibition Hall B2-B3, stand # 507
Curious about what Time Series Foundation Models actually learn? Stop by our poster today at #ICML2025! Presented by @MononitoGoswami and myself! 📍West Exhibition Hall B2-B3, stand #507.
Curious about what Time Series Foundation Models actually learn? Stop by our poster today at #ICML2025! Presented by @MononitoGoswami and myself! 📍West Exhibition Hall B2-B3, stand #507.
blog.ml.cmu.edu/2025/07/08/car… Check out our latest post on CMU @ ICML 2025!
TabM now has a Python package! TabM is a simple and powerful DL architecture for tabular data that efficiently imitates an ensemble of MLPs 🏆 TabM has been used in winning solutions on Kaggle, and performs well on TabReD -- a challenging benchmark! 💻 pip install tabm 👇Link
It is easier and flashier to evaluate LLMs on clean data like NEJM cases, but we can't start talking about "medical superintelligence" until we engage with the messy reality of actual real-world clinical data
Wow. Bullish on AI for clinical reasoning but nejm cases are not real world :) furthest thing from it highly curated, highly packaged none of my patients come with pithy blurbs distilling hours of conversations & chart reviews into pertinent positives and negatives
Presenting DemoDiffusion: An extremely simple approach enabling a pre-trained 'generalist' diffusion policy to follow a human-demonstration for a novel task during inference One-shot human imitation *without* requiring any paired human-robot data or online RL 🙂 1/n