وبلاگ تحقیقات هوش مصنوعی

Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory. Existing visual-token re...

Are AI agents tools, co-authors, or researchers? We present a quantified case study ($N=1$): a physicist supervising an AI coding agent (Claude Code, Sonnet and Opus models) over 12 work days and 57 s...

Many employers screen job applicants with algorithms built by the same few algorithm vendors. We hypothesize that algorithmic monoculture leads to the same individuals and members of the same racial g...

Large language model (LLM) agents rely on reusable skills to solve complex tasks. However, existing skill creation approaches treat skills as isolated and static artifacts, limiting their reusability,...

We present MobileGym, a browser-hosted, lightweight, fully controllable environment for everyday mobile use, targeting interaction fidelity without replicating proprietary backends. It enables two cap...

This paper studies the next major bottleneck in agentic AI as system scaling, not only model scaling: the design of auditable, persistent, modular, and verifiable architectures around foundation model...

Subject-driven image generation aims to synthesize new images that preserve the identity of the given subject while following textual instructions. Existing approaches often encode text and reference ...

Code review is a critical practice in software engineering, yet the growing scale and frequency of code patches in modern projects, together with the widespread adoption of AI code assistants, make ma...

اوپن‌ای‌آی در ۲۳ آوریل ۲۰۲۶ از GPT-5.5 به عنوان «هوشمندترین و شهودی‌ترین مدل خود تا امروز» رونمایی کرد. این مدل نشان‌دهندهٔ یک تغییر راهبردی بنیادین است: دیگر یک «مشاور پرحرف» نیست، بلکه یک «مجری عامل...