Natural Language Processing news and analysis on Web Pulse: 3952 curated articles tracking the latest Natural Language Processing developments, tools, and research, updated continuously from vetted sources.
Google Earth on the web is rolling out experimental Gemini-powered geospatial capabilities, including Ask Google Earth, a chat interface for natural-language map analysis, and Imagery search, a multimodal feature for dis…
Smallest.ai, a startup founded in late 2024, has raised $13 million in a Series A round led by Seligman Ventures with participation from Sierra Ventures and 3one4 Capital, bringing its total funding to over $21 million. …
Few-shot prompting with well-chosen examples outperformed larger models with zero examples in AI translation tests, cutting terminology errors from 9% to 2% on a 500-string test set. The technique, tested against DeepL, …
AssemblyAI's 2026 guide to building voice-controlled AI agents identifies orchestration—latency, turn-taking, tool calls, and interruption handling—as the core engineering challenge, not the prompt or model. The article …
Guardrails AI has launched the Guardrails Hub, a searchable repository of 70 validators designed to ensure the safety, quality, and compliance of AI-generated text. The hub includes validators for detecting PII, profanit…
A study using the HealthContradict benchmark found that standard biomedical retrieval-augmented generation (RAG) systems fail to surface contradictions in 72.6% of queries, despite high citation accuracy. The issue is st…
Zernote, a startup founded by Miron Beneval, launched Zernote Live and Zernote MCP, tools that help businesses analyze user interviews. Zernote Live suggests follow-up questions during calls and tracks conversation drift…
Alibaba's Qwen team announced on July 31st the release of Qwen-Audio-3.0-ASR-Flash, a speech-recognition service designed to improve domain-term recognition and structured transcript generation. In internal benchmarks, Q…
A developer built two open-source packages, micromark-extension-math-extended and remark-math-extended, to let Markdown parsers handle TeX-style math delimiters (\(...\) and \[...\]) commonly emitted by LLMs. The project…
PolyAI released Dialog-RSN-1, an audio-native dialog model that processes caller audio directly, integrating turn-taking, speech recognition, function calling, and response generation into a single model while keeping te…
Researchers introduced a benchmark for reasoning over probability operators, containing 14,320 procedurally-generated English prompts across fifteen inference templates, and found that most of the 29 evaluated large lang…
Researchers propose Conditional Retrieval Alignment (CoRA), a gradient-free framework that converts a frozen encoder into a task-conditioned retriever for on-device in-context learning, using closed-form ridge regression…
Researchers propose SentiLLM, a framework that uses Semantic-Aligned Structural Abstraction to improve multimodal sentiment analysis by converting non-verbal signals into text-like tokens for large language models. The m…
Researchers introduced OVEarth-Bench, a benchmark for open-vocabulary Earth observation that expands category breadth and query diversity, finding that current methods perform limitedly, with MLLM-based methods achieving…
A new arXiv preprint (2607.27566v1) finds that direct answer-only supervised fine-tuning (SFT) on MedGemma-1.5-4B is the most robust adaptation method for multi-frame medical visual question answering on the MedFrameQA b…
Researchers developed an LLM pipeline that converts audio interviews into transcripts, maps them into the ten MADRS symptom items, estimates severity, and identifies problematic clinical ratings, achieving a correlation …
Researchers propose Generator-Gate-Corrector (GGC), a framework for reliable LLM-based Text-to-SPARQL generation, which improves query-level accuracy from 90.23% to 98.33% on MCQA while reducing inference overhead by 45%…
Researchers from an unnamed institution proposed a structured information extraction framework using LLaMA-3.1-70B to extract six semantic dimensions from financial news, finding that combining these with FinBERT sentime…
Researchers present B1ade, an efficient retrieval-augmented generation (RAG) architecture with a 335M parameter embedding model and a 1B parameter small language model (SLM). B1ade-embed achieves top MTEB scores among su…
Researchers introduced MiGUE-Bench, a benchmark for evaluating large language models (LLMs) on multi-granularity event analysis, covering tasks from event detection to future prediction. The benchmark uses an LLM-driven …