All current LLM serving systems place the GPU at the center, from production-level attention-FFN disaggregation to NVIDIA's Rubin GPU-LPU heterogeneous platform. Even academic PIM/PNM proposals still treat the GPU as the central hub for cross-device communication. Yet the GPU's compute-rich architecture is fundamentall...
Zhongkai Yu, Haotian Ye, Chenyang Zhou et al.· 0 citations
The standard recipe for training reasoning models---SFT on rationale annotations followed by RLVR---relies on rationale annotations being available and clean. When no annotated rationales exist, SFT cannot start at all, and when the model is too unaligned, RLVR alone stalls at cold start; when annotations are weak, SFT...
Large language model agents rely on a memory harness to write, organize, retrieve, and use past experience. A harness that works for one task often fails on another, because conversations, embodied planning, and expert reasoning require different storage and retrieval behavior. To address this limitation, we introduce...
Wenbo Pan, Shujie Liu, Xiangyang Zhou et al.· 0 citations
Machine-learned interatomic potentials (MLIPs) are deployed for high-throughput materials screening without formal reliability guarantees. We show that a single MLIP used as a stability filter misses 93% of density functional theory (DFT)-stable materials (recall 0.07) on a 25,000-material benchmark. Proof-Carrying Mat...
Generalized additive models (GAMs) offer high interpretability through independent univariate feature effects but underfit when interactions are present in data. GA$^2$Ms add selected pairwise interactions which improves accuracy, but sacrifices interpretability. We propose Conditionally Additive Local Models (CALMs),...
Vasilis Gkolemis, Loukas Kavouras, Dimitrios Kyriakopoulos et al.· 0 citations
This paper introduces a novel recovery-based shielding framework that enables safe RL with a provable safety lower bound for unknown and non-linear continuous dynamical systems.
Alexander W. Goodall, Francesco Belardinelli· Adaptive Agents and Multi-Ag...· 2 citations
TIRx is presented, an ML compiler for GPUs and AI accelerators that exposes these ingredients through explicit interfaces and introduces tile-primitive dispatch, which lowers high-level operations to efficient implementations according to scope, layout, and backend.
Bo-Han Hou, H. Jin, Guan-Jie Wang et al.· 6 citations· ⚡2
Understanding the structure of complex, nonstationary, high-dimensional time-evolving signals is a central challenge in scientific data analysis. In many domains, such as speech and biomedical signal processing, the ability to learn disentangled and interpretable representations is critical for uncovering latent genera...
Ioannis N. Ziogas, Aamna Al Shehhi, Ahsan H. Khandoker et al.· 0 citations
Multimodal large language models (MLLMs) have demonstrated remarkable capabilities across vision-language tasks, yet their large-scale deployment raises pressing concerns about memorized private data, outdated knowledge, and harmful content. Existing unlearning approaches for MLLMs typically adapt training-based strate...
Chenlu Ding, Jiancan Wu, Leheng Sheng et al.· 0 citations
A rule-to-image synthesis pipeline that extracts and expands puzzle rules from seed questions and generates executable rendering programs for puzzle assembly to tackle the challenge of image synthesis with grounded answers is proposed.
Offline RL (ORL) promises safe and sample-efficient deployment but existing methods rely on undocumented online interactions for hyperparameter tuning and lack reliable fully offline estimates of initial online performance. We introduce SOReL, a fully offline Bayesian model-based RL method that learns a posterior over...
Mattie Fellows, Clarisse Wibault, Uljad Berdica et al.· 0 citations
Bayesian optimization (BO) is a sequential decision-making tool widely used for optimizing expensive black-box functions. Recently, Large Language Models (LLMs) have shown remarkable adaptability in low-data regimes, making them promising tools for black-box optimization by leveraging contextual knowledge to propose hi...
Chih-Yu Chang, Milad Azvar, Chinedum Okwudire et al.· 0 citations
Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility—and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.
Learn how Google Antigravity and our Cheminformatics Skills accelerate drug discovery workflows, search compounds, and propose novel rigidified macrocycles.
MIT News · Artificial Intelligence· news.mit.eduOct 28, 2026
Professor Paris Smaragdis develops computers with superhuman hearing for applications like speech recognition, music understanding, and environmental monitoring.
MIT News · Artificial Intelligence· news.mit.eduOct 27, 2026
Prabha Jha used her experience with MIT Open Learning’s program for senior executives to develop AI solutions for her employer, Verizon, and to write a book.
