Natural Language Processing news and analysis on Web Pulse: 3990 curated articles tracking the latest Natural Language Processing developments, tools, and research, updated continuously from vetted sources.
Ford Pro has introduced Ford Pro AI across Europe, a system that lets fleet managers ask questions in natural language and receive answers from connected-vehicle data, after research found fleet managers spend up to 20 h…
A developer designing AI voice agents for electricians emphasizes that the hard part is not making them talk but making them behave like a reliable front desk. The workflow uses a separate policy layer for deterministic …
A developer outlines a practical checklist to make websites visible to AI search systems like ChatGPT, Perplexity, and Gemini. The guide covers three key fixes: unblocking AI crawlers in robots.txt, implementing an llms.…
A new study finds that denoising audio with SAM-Audio preprocessing paradoxically worsens zero-shot automatic speech recognition (ASR) performance. Researchers tested five OpenAI Whisper variants on noisy Bengali and Eng…
A developer built a production-ready semantic search engine for e-commerce using open-source tools: sentence-transformers for embedding, FAISS for vector indexing, and FastAPI for serving. The pipeline transforms product…
Google is preparing a new AI feature for Google Photos that will let users edit photo metadata using natural language commands through the Ask Photos service, according to findings from Android Authority. The feature, st…
Audio-language embedding models like CLAP fail to understand negation, causing performance to drop below chance on tasks that require identifying the absence of a sound, according to a new evaluation framework called Neg…
Researchers have introduced a novel Audio-Visual Speech Enhancement (AVSE) framework that uses a Large Language Model (LLM) to generate narrative descriptions of enhanced speech, which are then analyzed by a sentiment an…
Large language models like OLMo-2 7B exhibit left-right brain asymmetry in language processing that mirrors the human brain's left-hemisphere dominance, according to research using fMRI. As the models train, their intern…
WordPiece, Unigram, and SentencePiece are three subword tokenization algorithms that differ in how they split text, with WordPiece using a likelihood-maximizing merge criterion and a ## continuation marker, while Sentenc…
Byte-Pair Encoding (BPE), originally a data compression algorithm published by Philip Gage in 1994, was adapted in 2016 by Sennrich, Haddow, and Birch for subword tokenization in neural machine translation, solving the v…
A language model does not see letters or pixels; it sees a sequence of integers from a vocabulary table, and the raw input is just bytes. Unicode assigns every character a unique integer code point, but encoding these co…
The encoder-decoder (seq2seq) architecture, proposed by Cho et al. and Sutskever et al. in 2014, compresses the entire source sentence into a single fixed-length vector, causing performance to degrade sharply on longer s…
Recurrent neural networks (RNNs) process sequences by maintaining a hidden state that compresses the entire history into a fixed-size vector, as formalized in Elman's 1990 paper "Finding Structure in Time." The hidden st…
GloVe, FastText, and word2vec are all word embedding methods that produce dense vector representations of words, but they differ in their underlying approaches: GloVe uses a weighted least-squares regression on global co…
Google researchers led by Tomas Mikolov published word2vec in 2013, a method that made high-quality dense word embeddings cheap to train at billion-token scale on a single machine, transforming them from a research curio…
N-gram language models suffer from the curse of sparsity: as context length increases, the number of possible n-grams grows exponentially (Vⁿ), far exceeding any training corpus, leading to most unseen n-grams receiving …
A language model assigns a probability to the next token given all previous tokens, a task that the chain rule shows is sufficient to capture any pattern in language. Claude Shannon demonstrated in 1951 that humans can p…
Language models generate text by constructing a categorical probability distribution over their vocabulary at each decoding step, converting raw neural network outputs called logits into valid probabilities using the sof…
A blog post explains that the L2 norm measures a vector's magnitude as its Euclidean distance from the origin, while the dot product quantifies similarity between vectors and is the core operation behind attention mechan…