AI News

Object removal models have improved faster than the metrics used to judge them. Diffusion erasers now reconstruct shadows, reflections and occluded structure convincingly, yet PSNR, SSIM, LPIPS, ReMOVE and CFD frequently rank their outputs the wrong way. The root cause is structural: erasure is an ill-posed, one-to-many task, so no single ground truth exists to…

import os, sys, io, json, time, math, re, subprocess, warnings from collections import Counter, defaultdict warnings.filterwarnings("ignore") os.environ.setdefault("USE_TORCH", "1") def _pip(*pkgs): subprocess.run([sys.executable, "-m", "pip", "install", "-q", *pkgs], check=False) try: import doctr except ImportError: print(">> Installing python-doctr (this takes ~1-2 min on Colab)...") _pip("python-doctr[viz]") try: import reportlab except ImportError: _pip("reportlab") import numpy as np import…

Yesterday, Liquid AI released LFM2.5-VL-3B. It is a 3.1B-parameter vision-language model built for on-device deployment. The model reads digital screens across mobile, web, and desktop. It grounds objects to coordinates, parses documents and charts, and calls tools from text or image input. Liquid AI reports an average of 69.4 across 28 vision benchmarks. That matches…

In this tutorial, we explore adaptive experimentation using Meta’s Ax with the modern Client API. We work through a complete workflow where we tune a RandomForest model on a synthetic classification dataset while balancing predictive accuracy against model footprint. We begin by defining a mixed search space with integer, float, log-scaled, and categorical parameters, then…

Onton, a San Francisco-based search and discovery company, has released Ontology 1, a neurosymbolic model for complex, conversational, multimodal product search. On a 90-query benchmark scored by three independent LLM judges, Ontology 1 reached a mean precision@10 of 0.630, against 0.543 for Google Shopping and 0.469 for Amazon. It did this while indexing roughly 1%…

AI developers, researchers, and professionals frequently hit a frustrating wall when analyzing large documents with LLMs: the hidden, compounding cost of context windows. Pasting a 200-page PDF into a chat isn’t a one-time charge. Because the conversation history is re-sent to the model on every single turn, that massive document is paid for again with…

NVIDIA just released DeepStream 9.1. The update targets a persistent problem in video analytics. Tracking one object across many cameras traditionally requires manual camera calibration and complicated calculations. DeepStream 9.1 addresses this with two additions: Multi-View 3D Tracking (MV3DT) and AutoMagicCalib (AMC). Both ship as agentic skills for coding agents. As a result, developers move…

import torch.nn as nn import torch.nn.functional as F from dataclasses import dataclass torch.manual_seed(0) @dataclass class Cfg: d_model: int = 192 n_head: int = 6 n_layer: int = 4 ffn_mult: int = 2 n_mod: int = 3 …

def _purge(*prefixes): for name in [m for m in list(sys.modules) if any(m == p or m.startswith(p + ".") for p in prefixes)]: del sys.modules[name] def _load_ocrmypdf(): _purge("PIL", "ocrmypdf") import…

Datalab has released lift, a 9B open-weights vision model for structured extraction. You pass it a JSON schema, and it returns a JSON object that matches. The model reads PDFs and images directly, then decodes against your schema. This is Datalab’s first model built purely for extraction. The team already ships open-source OCR tools: chandra,…