RecommenderSystems
Most apps show a fixed number of “top” items—say 10 movies or 20 products—assuming there are always enough good options. But that’s not always true: sometimes there are few relevant items, or some users are extra picky. The result? Filler recommendations. Dynamic-K flips the script. Instead of
AI
Teaching chatbots to stop contradicting themselves Ever had a bot say one thing, then the opposite a few turns later? This study introduces DECODE—a new task and dataset for spotting contradictions in everyday conversations, drawn from both human-human and human-bot chats. * New data beats existing natural language inference (NLI)
AI
Nature builds brains from a single cell. This study shows how an artificial network can do something similar - growing itself from "noise". Inspired by the early visual system (retina to LGN), the authors propose a simple developmental algorithm that starts with one "cell" and self-organizes
ReinforcementLearning
Reinforcement learning (RL) wins in games and simulators—but deploying it on real products is a different story. Gabriel Dulac‑Arnold, Daniel Mankowitz, and Todd Hester outline nine must-solve challenges before RL can safely power real-world systems. * Safety & constraints: avoid harmful actions while learning. * Sample efficiency: learn from limited,
MachineLearning
Training AI usually needs lots of human-labeled examples. This work shows a different path: make models learn from plenty of unlabeled data by asking them to stay consistent under strong "noise." Instead of simple tweaks (like small crops or word drops), they use powerful data augmentation: RandAugment for
AI
Can AI essay graders be fooled? New research shows many automatic essay scoring (AES) systems reward “word soups” over real writing. Using interpretability tools, the authors found that a few keywords drive most of the score, while flow, grammar, and coherence matter far less. The wild part: because these models
AI
Meet AdaRec: smarter recommendations from fewer clicks AdaRec is a new way to personalize recommendations using large language models. Instead of hand-crafted features, it turns your past interactions into short, readable stories—“narrative profiles”—and lets an LLM reason over them with just a few examples. It thinks in two
AI
Large language models can sound confident even when they are wrong. This study pinpoints a sneaky culprit: spurious correlations — statistical shortcuts like linking certain surnames to a nationality. Models absorb these patterns from data and then answer with high certainty, even when the link is false. What the researchers found
AI
Imagine telling your phone what you want—and an AI taps, types, and swipes to get it done. This paper introduces DigiData, a large, diverse, multi‑modal dataset designed to train general‑purpose mobile control agents. Instead of relying on unstructured user logs, the team systematically explores app features to
AI
TL;DR LoRA adapters are small plug-in modules that fine-tune big language models cheaply. But one adapter per task doesn’t fit messy, real-world inputs. What’s new LoRA on the Go (LoGo) lets a model pick and blend the best adapters for each individual input—no labels, no extra
AI
AI finally masters Stratego—on a budget Stratego has long been a worst-case test for AI: you must plan deeply while most pieces are hidden. Big-budget attempts still fell short of top humans. This paper reports a step change: an AI that reaches vastly superhuman Stratego performance—trained for only
AI
Big win for AI in hidden-information games. Stratego has long resisted top-tier AI—even with training runs costing millions. This new study cracks it. Researchers achieved vastly superhuman Stratego play using self-play reinforcement learning plus a smart test-time search designed for imperfect information. The surprise: it worked with only a