AI·Virginia Tech researchers found that resistance training (modeled as mouse weightlifting) outperforms running for reducing fat and improving insulin sensitivity and glucose control in obesity and Type 2 diabetes models.
Search
What are you looking for?
60 results for “nn”
AI·The iroh.computer blog introduces Mesh LLM, a distributed AI computing system running on the iroh network protocol.
AI·SQLite strict tables enforce column types to prevent datatype errors like inserting text into integer columns, allowing only standard types or ANY for flexibility, but cannot be added to existing tables and require SQLite 3.37+.
AI·ClickHouse scaled PgBouncer throughput 4x by running a fleet of processes sized to cores, using so_reuseport to balance connections on one port and inter-process peering for query cancellation.
AI·Blackwell RTX Spark is demonstrated on a real machine at Bilibili World, with CPU and GPU directly soldered together for running 120B-parameter models on laptops.
AI·Zhipu AI founder Tang Jie releases internal letter after 'GLM Moment' success, announcing focus on long-horizon tasks, autonomous agents, and self-evolving AI beyond coding and reasoning.
AI·SpaceX filed FCC for 100,000 Gen 3 Starlink satellites with 10x bandwidth upgrades, explicitly linking the constellation to AI workloads, data centers, autonomous vehicles, and humanoid robots for global compute connectivity.
AI·A user shares benchmarks of dual RTX 3090 and Titan RTX cards running llama.cpp with Qwen3.6-27B at 180k context, comparing tensor and pipeline parallel modes for PCIe transfer efficiency.
AI·Quantum Computing China has completed its first 100,000-card cluster, achieving national full-stack computing power support for the ten-thousand-card era and successfully running over 300 applications.
AI·The post shares practical recipes and benchmarks for running LLMs on popular consumer hardware, with a new website listing hardware-filtered setups and user-voted usage stats.
AI·VultronRetriever family of models is released on Hugging Face, topping MTEB leaderboard classes including VultronRetrieverPrime-8B as global #1, with offline iPhone Q&A and embedding demo.
AI·LLM-as-judge scores fluctuate across evaluators even with fixed candidate responses, treating the issue as measurement validity. Across four datasets, scaling Qwen3 from 1.7B to 32B parameters yields only limited adjacent gains, while MiniMax M2-to-M2.7 API upgrades show none; stronger judges reduce but do not eliminate position and verbosity biases, with repeated juries offering little benefit under correlated errors.
AI·User with RTX 4090, 3090Ti, and 128GB RAM struggles to run larger models like 122B due to VRAM limits while Qwen3.6-27B runs efficiently without using system RAM.
AI·Viral stories of AI-crafted mRNA vaccines saving pets or spotting cancers mirror the Theranos blood-testing fraud, with exaggerated claims by non-experts creating hype while AI merely integrates existing tools rather than innovating breakthroughs.
AI·This position paper reviews LLM-driven formal theorem provers and argues that current systems function mainly as solvers for well-defined problems, not as research agents capable of discovering new theorems or resolving open conjectures at the frontier. It identifies key limitations in datasets, exploration, tools, and collaboration and proposes a roadmap for AI4Math systems to support genuine mathematical research.
AI·Hidden Decoding expands each token into n independent streams during continued pretraining of a fixed Transformer backbone, using separate embedding tables and retaining intermediate KV caches for latent computation; Stream-Factorized Attention limits cross-stream mixing to reduce costs. Experiments show frontier 80B and 617B MoE models improve on all benchmarks over matched baselines.
AI·LLM-based ASR in regulated domains like banking is limited by privacy and real-speech collection costs; synthetic TTS data is a cost-effective substitute, but acoustic mismatch hinders supervised fine-tuning (SFT). Group Relative Policy Optimization (GRPO) applied solely to synthetic speech reduces WER by 40% relative to SFT (36.71% to 22.09%) and by 45% in the SFT-then-GRPO sequence, by improving behavioral calibration and audio attention rather than representations.
AI·Across five instruction-tuned LLMs, ordinary answer-confidence separates correct from wrong answers on answerable questions but fails to distinguish unanswerable questions (e.g., false-premise ones in CREPE), while a linear probe on hidden states performs the opposite, revealing two distinct abstention axes that remain separable even at 14B scale.
AI·DominoTree is a training-free best-first tree draft builder for speculative decoding that scores each candidate node by re-applying Domino's GRU-based causal correction along its specific root-to-node path. It restricts per-node correction to top-M candidates for efficiency and delivers up to 6.6x speedup with the highest mean accept length of any tested method.
AI·MASTE: A Multi-Agent Pipeline for Zero-Shot Aspect Sentiment Triplet Extraction
AI·ICDAR 2026 HIPE-OCRepair competition evaluated LLM-assisted post-correction of noisy OCR from 17th-20th century multilingual (EN/FR/DE) historical newspapers and books. Four teams used zero-shot to fine-tuning approaches; results show significant error reduction but recurring over-correction on low-noise inputs, with a public dataset and evaluation framework released.
AI·XALPHA is a memory-driven AI quant researcher that uses multi-source research memory integrating external financial reports and prior discovery feedback, with Macro Brain for theme planning and archetype selection, Micro Brain for hypothesis-to-code translation and tri-alignment verification, and Cross Brain for feedback consolidation, enabling closed-loop continuous alpha discovery that outperforms baselines on CSI300.
AI·Achieves 50-54 tok/s for Qwen3 30B A3B float8 on 16GB RTX 5060 Ti using custom CUDA/C++.
AI·The paper introduces Hallucination Self-Play, a bootstrapping approach that uses an evolved generator to reinforce a detector for identifying faithfulness hallucinations in LLMs.
AI·The paper introduces a graph-based framework to quantify uncertainty, coherence, and robustness in LLM reasoning, addressing gaps in decoding strategies like Self-Consistency that only check final-answer agreement.
AI·SQuaD-SQL uses LLM-guided synthetic data and LoRA fine-tuning to train 1.5B-parameter SLMs to reach 86.9% execution accuracy on WikiSQL, matching large models while requiring only one consumer GPU and delivering faster, lower-memory inference.
AI·Researchers developed SQZ-Qwen-ASR-1.7B for MLC-SLM 2026 Task 1, combining a modular speaker diarization front-end (VAD, CAMPPlus embeddings, spectral clustering, RTTM segmentation) with Qwen3-ASR-1.7B adapted via full supervised fine-tuning, LoRA on TTS-generated synthetic speech, and GRPO reinforcement learning for lower tcpMER (23.70 on dev set, 17.97 on eval set).
AI·The paper proposes UltraX, a method for refining large-scale pre-training data using adaptive programmatic editing to improve LLM quality when scaling data yields diminishing returns.
AI·The Reddit post explores how the context-aware viewpoint of neural networks leads to a simplified view of layers as average best linear mappings.
AI·User compares 5090 at 600W full compute and prompt processing vs shunt-modded 6000 PRO MaxQ water-cooled at 300-600W.
AI·Discusses feasibility of running Qwen 122B on 64GB RAM + 24GB VRAM and suitable settings.
AI·DeepSearch-World is a self-distillation framework for web search agents that uses self-generated experience for training in verifiable environments.
AI·The research finds that preprocessing-based stereotype mitigation in NLP can backfire by increasing stereotyping or counter-stereotyping for some groups relative to neutral baselines.
AI·The paper identifies a failure mode in critic-free RL methods for LLMs where implausible tokens receive uniform credit, and proposes tail-aware credit calibration to improve reinforcement learning.
AI·The paper empirically assesses the reliability of Gemini models as audio judges for full-duplex voice agents by scoring stereo waveforms, validated against human calibration.
AI·The paper proposes an agentic tool-making pipeline that compiles repeated SOP steps into validated versioned tools to reduce latency and improve reliability in production LLM agents.
AI·The paper introduces PLURAL, a large-scale value-focused preference dataset grounded in the Integrated Values Survey across 92 countries to improve LLM representation of diverse non-Western value systems.
AI·Holographic Neural PCFG induces latent constituency trees from raw text using holographic memory and neural rule scoring for unsupervised parsing.
AI·Prompt Compression via Activation Aggregation compresses task-relevant prompt information into a single activation vector for re-injection into the model.
AI·Formal verification method using ESBMC-PLC+ detects Ladder Logic Bombs in IEC 61131-3 PLC programs by focusing on function-block bodies where malicious code hides.
AI·Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoders extract cross-seed universal features from independently trained BERT models to address dictionary learning misalignment in mechanistic interpretability.
AI·Structured pruning method for sparsely-activated MoE models removes bad experts to address full expert bank memory bottleneck while preserving inference efficiency.
AI·COALA enhances contextualized SLMs for multi-entity ASR via contrastive regularizer and biasing score estimation to handle domain-specific entities robustly.
AI·Grounded Event Extraction from SEC 8-K filings uses a fine-grained taxonomy to overcome coarse SEC item codes for market-moving disclosures.
AI·The paper presents a cost-efficient human-LLM collaborative annotation framework to construct the EspanSt stereotype dataset for non-English languages and underrepresented cultures.
AI·The paper introduces a multi-cluster boundary learning method using MiniLM embeddings for detecting out-of-scope intents in human-machine interaction systems.
AI·The paper introduces a Bloom-aligned framework to measure educational control in LLMs, focusing on preserving instructional intent while aligning cognitive demand with learning objectives in programming tasks.
AI·The paper proposes structured pruning for LLMs using power transformation and sign-preserving score aggregation with adaptive feature retention to address distribution mismatch issues.
AI·The paper probes internal representations of Eternis-Forecaster 8B for forecasting, training a representation-pooling method to assess calibration and faithfulness of internal CoT reasoning.
AI·COBART uses a bidirectional auto-regressive transformer for optimized ad headline generation with multi-objective control over quality and CTR.
AI·TypeProbe recovers type representations from hidden states of pretrained code models using parallel Java and Python datasets.
AI·LEXIC pushes gaze-only reading comprehension prediction on EyeBench with lightweight language-model-free conditioning and injected complexity.
AI·Best-of-N TTS evaluation is confounded by ASR family alignment, where verifier quality depends on the judging ASR model family.
AI·Large-Language-Models-as-a-Judge enable theory-agnostic adaptive metric-alignment for prototypical networks in personality recognition.
AI·The study applies LSTM and traditional models to analyze public sentiment on social media platforms like Twitter regarding real-time events and issues.
AI·CKTN multilingual corpus covers Cham, Khmer, and Tay-Nung from Vietnam's highlands, delta, and coast for NLP of under-resourced minority languages.
AI·The paper enriches Roy Harris's Integrationism theory with Barenholtz's Autogenerative Theory to address gaps in computational language approaches.
AI·The third Ant InTech Award submissions deadline approaches in seven days, with Turing Award winners on the judging panel and awards available in four directions.
AI·OpenAI announces the GPT-5.5 Bio Bug Bounty program with details of the new bounty initiative.
AI·OpenCoF enables reasoning in video generation through temporal frame chains distinct from Chain-of-Thought, addressing training data gaps.