Internal Pluralism and the Limits of Pairwise Comparisons
arXiv:2607.02672v1 Announce Type: new
Abstract: Local pairwise comparisons are a standard tool for learning how people want decision rules to work, e.g., in participatory design or alignment. However, their use builds in two strong assumptions: that local comparisons are sufficient evidence about h...
ASK in the Dark: Uncertainty-Gated LLM Assistance under Partial Observability
arXiv:2607.02686v1 Announce Type: new
Abstract: Reinforcement learning agents operating under partial observability must act on incomplete information, making them natural candidates for guidance from small language models (SLMs) that carry broad reasoning priors. Yet integrating SLM guidance into ...
arXiv:2607.02771v1 Announce Type: new
Abstract: Leadership computing facilities steward large-scale scientific datasets that routinely require substantial transformation before serving as AI training data. However, no existing framework fully unifies automated transformation, readiness assessment, ...
SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery
arXiv:2607.02807v1 Announce Type: new
Abstract: Long-running coding agents such as autoresearch can persistently discover optimizations for open-ended problems. However, they tend to converge onto a single high-level approach, then proceed with low-level edits while missing other superior approache...
Building a Scaffold-Split Random Forest QSAR Co-Scientist for EGFR Inhibitor Discovery Using ChEMBL, RDKit, SHAP, and BRICS
In this tutorial, we build an autonomous AI co-scientist for EGFR C797S inhibitor discovery. We resolve the target through ChEMBL and UniProt, then mine IC50 records into a clean pIC50 dataset. We use RDKit to standardize molecules, compute Morgan fingerprints, and train a scaffold-split Random Fore...
Taming Text-to-Sounding Video Generation via Advanced Modality Condition and Interaction
This study focuses on Text-to-Sounding-Video (T2SV) generation, which aims to generate a video with synchronized audio from text, with both modalities aligned to the text conditions. Despite progress in joint audio-video training, two critical challenges remain: (1) text conditioning is a bottleneck...
DynaMiCS: Fine-Tuning LLMs with Performance Constraints Using Dynamic Mixtures
Multi-domain fine-tuning of large language models requires improving performance on target domains while preserving performance on constrained domains, such as general knowledge, instruction following, or safety evaluations. Existing data mixing strategies rely on fixed heuristics or adaptive rules ...
LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
Vision Language Models (VLMs) offer the exciting possibility of processing text as rendered images, bypassing the need for tokenizing the text into long token sequences. Since VLM image encoders map fixed-size images to a fixed number of visual tokens, varying rendering resolution provides a fine-gr...
MT-EditFlow: Reinforcement Learning for Multi-Turn Image Editing with Flow Matching
Recent breakthroughs in instruction-based image editing have captured significant attention, as models are now capable of handling real-world editing demands with the practicality required by everyday users. However, editing models trained primarily for single-turn edits often break down in multi-tu...
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
The web is complex, open-ended, and constantly changing, making it challenging to scale training data for visual web agents. Existing data collection attempts remain limited to offline trajectories for supervised fine-tuning or a handful of simulated environments for RL training, thus failing to cap...
The ‘first’ AI-run ransomware attack still needed a human
An AI agent carried out the technical execution of a real-world ransomware attack for the first known time, but new details show a human still chose the victim, set up the infrastructure, and supplied stolen credentials — meaning it wasn't quite the fully autonomous cybercrime debut that last week's...
Clash of the Digital Titans: How Gemini and GPT-4o View the Global Trade Crisis
What happens when Google’s Gemini and OpenAI’s GPT-4o debate global trade? Discover how the world's top AI models analyze supply chain chaos differently.
Every year, the International Conference on Machine Learning (ICML) reveals where thousands of AI researchers have decided to put their work. This year’s accepted papers reveal a clear direction: open frontier models and open AI infrastructure have become foundational to how modern AI science gets ...
5 Ways Small Language Models Are Powering Next-Gen Agents
This article looks at five concrete ways SLMs are showing up inside next-generation agents right now, from the research backing them to the tools and numbers worth knowing if you're deciding whether your next agent needs a frontier model at all
Best-worst comparisons, MaxDiff-style judging, and Plackett-Luce utility scores give agent teams a cleaner way to decide which configs to ship, prune, and route toward next.
The post Stop Ranking Agent Configs by Average Score appeared first on Towards Data Science.
Sakana AI Launches Sakana Translate, a Namazu-Powered Japanese–English–Chinese Translation Tool With Translate, Proofread, and Ask Modes
Sakana AI has added Sakana Translate to Sakana Chat. It runs on the Namazu model series. The tool translates bidirectionally across Japanese, English, and Chinese. Three modes ship together: Translate, Proofread, and Ask. This article breaks down each mode, the XCOMET-XL benchmark, and practical use...
Modern VLMs Explained: How GPT-4o, Gemini, Claude Vision, and Qwen-VL Work
Vision Language Models, or VLMs, are AI models that can understand both visual content and language. While earlier models like CLIP and BLIP connected images with text, modern VLMs can analyze images, read documents, interpret charts, answer visual questions, and support multimodal conversations. Mo...