AI Research: Latest AI News and Analysis

Track AI Research updates with focus on product moves, market signals, and high-impact developments.

Latest Articles

• CRAF: Cross-View Residual-Aware Fusion for Deepfake Speech Detection — CRAF fuses self-supervised acoustic models with Auditory Large Language Models via residual-aware cross-view attention to detect deepfake speech, achieving 5.96% EER on ASVspoof 5—improving generalization to spoofing attacks not seen during training.
• LePlanner: An Iterative Amortized Controller For World Models — LePlanner, an amortized iterative controller for world model-based robot planning, matches search planners (CEM, MPPI) at 98% success on PushT and 100% on Reacher while requiring 3-49x less wall-clock time per decision—bringing fast goal-aware robot control closer to deployment.
• The Attribution-Compression Frontier in Retrieval-Augmented Generation — Abstractive context compression in RAG achieves 0.86 citation precision against summaries it creates but only 0.12 against original source spans — a 7x attribution gap that undermines citation faithfulness in any RAG system relying on compressed context for cited answers.
• Enemray: Toward Capable Language Models for Hassaniya — Enemray is the first Hassaniya-centric language model for the Arabic dialect spoken in Mauritania, achieving the strongest English-to-Hassaniya translation among tested open and proprietary models while retaining most general capabilities on math, code, and function calling.
• FedV-KGQA in Practice: Design Lessons and an Interactive Prototype — FedV-KGQA recovers most centralized accuracy on multi-hop knowledge graph QA without sharing raw triples between organizations — accepted as a poster at ISWC 2026 with an interactive prototype demonstrating full pipeline traceability.
• Hardware-Aware Learned Representation Compression for Distributed In-Sensor Vision — OASIS reduces in-sensor vision system energy by 2-4.5x through a lightweight encoder that compresses image representations before off-chip transmission, achieving 18,816x compression versus raw 8-bit pixels for visual wake-word detection on a Xilinx FPGA with under 1% accuracy loss.
• Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures — Continual Search, an iterative root-cause attribution framework for AI agent failures, improves GPT-5.5's F1 score on long-horizon failure diagnosis from 0.349 to 0.498—and shows that lower-tier models using effective search can surpass higher-tier models relying on one-shot judgment.
• Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself — Researchers at a large video streaming service fine-tuned an LLM recommender to generate personalized, faithful, and non-harmful explanations using constrained GRPO — raising the all-three-criteria pass rate from 65% to 96% without degrading recommendation quality.
• GeoSkill:Experience-Driven Hierarchical Skill Learning with Collaborative Revision forGeospatialAgents — GeoSkill proposes a hierarchical skill bank framework for geospatial agents that distills past execution experience into reusable planning and tool-use skills, with a multi-role revision mechanism that prevents error misattribution from polluting the skill store.
• IMM-based Multiple Object Tracking using a State Prediction Neural Network — A transformer-based radar tracking method (PR-IMM) integrates a neural state predictor into the classical Interacting Multiple Model framework, reducing position estimation error by 57.3% over standard IMM and cutting ID switches by 25.3% for autonomous vehicle object tracking.