GenerativeAI
Build stories like a graph What if writing felt like assembling LEGO? This research introduces a node-based editor that lets you craft multimodal stories—mixing text, images, audio, and video—one beat at a time. * Graph-first storytelling: Each scene is a node you can expand, swap, or refine with natural-language
AI
Ever read code and wonder "why does this exist"? This research taps into GitHub's natural-language breadcrumbs like pull requests, issues, and commit messages to help AI explain code with real project context. How it works: * Gather relevant GitHub discussions and structure them. * Explain the code'
AI
Encrypted LLM chats can still leak their topic Whisper Leak shows that even with TLS, streaming responses create telltale packet size and timing patterns. Across 28 popular models, these fingerprints let an eavesdropper classify prompt topics with near-perfect accuracy, flagging sensitive ones like "money laundering" and even recovering
GenAI
GenAI is reshaping genomics, proteomics, transcriptomics, structural biology, and drug discovery. This systematic review asks six big questions to map where GenAI helps most—and what’s next. * Where it works: Sequence analysis, molecular design, and integrative modeling often beat traditional methods via pattern discovery and realistic output generation. * Best
AI
AI agents promise to speed up discovery by doing the messy parts of research—forming hypotheses, writing code, running experiments, and analyzing results. But do they actually handle end-to-end projects? InnovatorBench is a new benchmark+platform that tests agents on realistic Large Language Model (LLM) research workflows. * 20 tasks across
AI
TL;DR We often know what we want, but struggle to tell AI. This “intention gap” wastes time. This paper reframes AI from a passive follower to a Socratic partner that asks the right questions. * What’s new? An agent called Nous learns an inquiry policy that maximizes information gained
AI
Interpreting what large language models “think” is slow and expensive. Each new model often needs custom tools and lots of manual labeling. Atlas-Alignment offers a shortcut. Instead of rebuilding everything, it lines up a new model’s hidden activity with a shared, human-labeled Concept Atlas using only overlapping inputs and
AI
How can AI agents and developers find API and data services they can actually trust—especially when anyone can spin up fake identities to game the rankings? David Shi and Kevin Joo propose TraceRank for the x402 network, where services accept crypto payments for access. Instead of counting raw traffic,
AI
AI agents are getting good at coding and research, but they still struggle with long, domain-specific projects. Training them is hard: full-time human annotation is too costly, while pure trial-and-error rarely finds successful paths. Apollo is a new sampling framework that blends lightweight, asynchronous human guidance with strict action-level filtering.
AI
If your AI segments tumors, who is it serving—and who might it be failing? A new audit of the MAMA-MIA breast cancer segmentation dataset examines fairness not in classification, but in segmentation—the step that often feeds every downstream decision. * Researchers evaluated automated segmentation quality by age, ethnicity, and
AI
From doodle to design Imagine sketching a rough poster layout and instantly getting a polished version. That’s the promise of Sketch‑to‑Layout, a new approach that uses your sketch as the design guide. * How it works: A multimodal transformer takes your quick sketch plus content (text/images) and
AI
What's new Spiking Neural Networks (SNNs) mimic brain cells that communicate with brief spikes, not continuous signals. That makes them naturally energy-efficient and great at timing. * Surrogate-gradient training gets SNNs within 1-2% of standard ANN accuracy, converging by about 20 epochs with latency near 10 ms. * ANN-to-SNN conversion