ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode

Search: “decoder”

145 stories in the last 30d

Searchkeywordsmart Sortlatesthotimportancerelevance
DomainAllCyberAIWindow24h3d7d30dAllIn the wildImportant onlyShow duplicates
CategoryAllAI research 47AI industry 43Model release 27Malware 20Vulnerability 18AI safety & security 16Exploit / PoC 10Research 8Phishing & fraud 7Threat actor 6AI tools & infra 6AI policy 5Industry 4Advisory 4
SeverityAllcritical 8high 31medium 30low 1info 155

A Princeton Researcher Proposes Recurrent Looped Transformer (RLT) that Carries Decoder State across Every Token, Fixing 96 Blocks per Token with Unbounded Temporal Depth

Princeton researcher Yifan Zhang proposes Recurrent Looped Transformer, carrying full decoder state across every token for unbounded temporal depth.

Yifan Zhang's technical report defines the Recurrent Looped Transformer (RLT), pairing a causal encoder with a recurrent decoder whose final output and layerwise sliding-window attention cache carry into every subsequent token with no prompt-response boundary reset. The reference configuration ties 48 encoder and 48 decoder layers, executing 96 logical blocks per token while the state path grows to 48t blocks after t tokens at fixed per-token compute. The report details RL replay contracts that rebuild all states under current parameters and exact prefix snapshots for multi-turn serving, but explicitly reports no measured efficiency, reasoning quality, or scaling results.

MarkTechPost · 2d agoAI research1

Speculative Decoding in vLLM on AMD GPUs

vLLM benchmarks speculative decoding on AMD Instinct MI300X and MI355X GPUs across five drafting methods including EAGLE-3 and native MTP.

The vLLM project documents draft-and-verify speculative decoding support for AMD GPUs via ROCm, comparing native MTP, Gemma 4 MTP, EAGLE-3, DFlash, and DSpark drafting approaches. Output-token throughput effects varied with drafting method, proposal length, model family, draft checkpoint, workload, and acceptance behavior. The post also covers how to enable each method plus practical tuning and observability considerations.

Hacker News · AI · 9d agoAI tools & infraHN 141↑ · 52 comments1

[AINews] DeepSeek v4.1-Flash: 763B-P8B-D16B novel causal Encoder–Decoder architecture with vision marks the Return of the Whale

DeepSeek released V4.1-Flash, an open-weight 763B-parameter model with a novel causal encoder-decoder architecture, 1M context, vision input, and MIT license.

DeepSeek launched V4.1-Flash, an open-weight MIT-licensed model using a novel causal encoder-decoder architecture with 763B total parameters and asymmetric active parameters: 8B for prefill and 16B for decode. It supports 1M-token context and text+image input, priced at $0.30 per 1M input and $1.20 per 1M output tokens with a 50% off-peak discount. Artificial Analysis scored it 40 on its Intelligence Index, above DeepSeek V4 Pro 0813, and Vals ranked it the #1 open-weight model ahead of Kimi K3. Baseten shipped day-0 support and Ollama began rolling it out to paid subscribers.

Latent Space · 4d agoModel release 7 sources1

Hardware Fingerprinting FTQC via Quantum Decoder Timing

Quantum decoder timing on IBM Heron processors forms a side channel enabling device fingerprinting with 89% accuracy and workload inference.

The work demonstrates that wall-clock syndrome-decoding times on fault-tolerant quantum computers constitute a novel hardware side channel. Using per-shot decoder timings from three IBM Heron processors collected over 68 days, a passive observer can reconstruct detector-firing distributions, estimate logical error rate, infer code distance, and fingerprint the specific device with up to 89% accuracy versus 33% for random guessing. Noisy simulation based on Google's 105-qubit Willow processor distinguishes nine surface-code patches at 81% accuracy, showing the channel persists across vendors and code families.

arXiv cs.CR · 5d agoResearch

Jackrong/Qwopus3.8-27B-Flash-GGUF — new model trending #26 on Hugging Face

Community fine-tune Qwopus3.8-27B-Flash, built on Qwen3.8-27B, cuts agent reasoning latency with 12.8% faster decoding and 80.7% MTP acceptance.

Jackrong released Qwopus3.8-27B-Flash, a fine-tune of Qwen3.8-27B optimized for long-running agent workloads, reporting 12.8% faster decoding and 80.7% multi-token-prediction acceptance. Training used roughly 1.5 million teacher-scored SFT examples filtered to the top 10%, followed by reinforcement training with NVIDIA NeMo-RL and GSPO. The author notes an explicit trade-off: MMLU-Pro mixed-set scores are lower than the base model, and a known bug can produce incorrect Python indentation. Author-provided benchmarks have not been independently verified.

Hugging Face trending models · 12d agoModel release1

ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation

ROSETTA is a hybrid CKKS/TFHE homomorphic encryption framework for privacy-preserving LLM decoding, achieving up to 4.8x Softmax and 2.1x end-to-end speedups.

The paper proposes ROSETTA, a hybrid CKKS/TFHE fully homomorphic encryption framework for private inference on generative LLMs, targeting the nonlinear operations that dominate autoregressive decoding cost. It introduces an adaptive segmented lookup-table protocol based on TFHE and a scheme-aware operator-selection framework that assigns each nonlinear operator to CKKS or TFHE to minimize latency. Experiments show up to 4.8x Softmax speedup and 1.5-2.1x end-to-end decoding speedup over the state-of-the-art CacheMir framework.

arXiv cs.CR · 1d agoResearch

TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimation

TransNormal-2 improves monocular surface-normal estimation by fixing VAE edge degradation with geometry-aware losses and refinement, matching MoGe-2 with 1.4% of annotations.

TransNormal-2 is a FLUX.2-based rectified-flow framework for monocular surface-normal estimation with single-step deterministic inference. The authors quantify that VAE 8x spatial compression introduces 1.3-8.5 degrees of mean angular error even on ground-truth normals, with edge error up to 2.8x the global error. The method adds geometry-aware pixel-space losses and an RGB-guided Geometric Refinement Module to correct boundary-localized decoding errors. It matches or exceeds MoGe-2 on all eight reported metrics using only 1.4% as many task-specific annotations, and cuts transparent-object MAE by 4.2 degrees on ClearGrasp and 3.1 degrees on ClearPose.

Hugging Face daily papers · 10d agoAI research

The Semantic Bottleneck: Leveraging Semantic Representations for Non-Invasive Speech Decoding

Researchers introduced Brain2Semantics2Text, decoding sentence meaning from non-invasive MEG brain recordings via a semantic bottleneck, improving on prior Brain2Text methods.

The paper proposes Brain2Semantics2Text, a non-invasive speech decoding method that maps sentence-level magnetoencephalography (MEG) responses into a semantic embedding space and inverts those embeddings into natural language. Motivated by evidence that high-level semantic representations are distributed across cortex and evolve on slower timescales, the approach targets meaning rather than phonemes or words, avoiding the need for word-level alignment. The authors report improved sentence-level results compared to prior non-invasive Brain2Text methods despite the low signal-to-noise ratio of neural recordings.

Hugging Face daily papers · 7d agoAI research2

Canonical Color as a Lens into Concept Decodability in Vision Encoders and VLMs

Probing study shows vision encoders make canonical color linearly decodable from grayscale images and tie it to object identity.

Researchers use canonical color as a controlled testbed for measuring conceptual (not just visible) information in vision encoder representations. A dataset of objects with canonical colors was built, and probes on both color and grayscale images show canonical color remains decodable even when color is removed from the input, linked to predicted object identity. Extending to full VLMs, they find post-training has a surprisingly large effect on color decodability in the vision encoder.

arXiv cs.AI / cs.LG / cs.CL · 7d agoAI research1

Has MIMO decoding been proved hard from lattice problems?

Researchers show the published lattice-hardness proof for MIMO decoding fails, as Regev's LWE reduction structure does not carry over to non-modular MIMO.

The paper re-examines Dean and Goldsmith's proposed polynomial-time reduction from lattice problems to MIMO decoding, which adapted Regev's reduction for learning with errors (LWE). Prior works had presented attacks and counterexamples against the construction, leaving the reduction's precise validity unclear. The authors identify which structural features of the LWE reduction fail to transfer to the non-modular MIMO setting, showing the published proof does not establish the claimed hardness of MIMO decoding. They distinguish flaws in the hardness proof from direct attacks on specific parameter choices and do not rule out physical layer security for MIMO systems in general.

arXiv cs.CR · 12d agoResearch

How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus

Reproduction study finds Orthrus speculative-decoding trajectories match the reference model in only ~45% of cases under BF16, but 100% under FP32.

Researchers independently reproduced Orthrus, a hybrid autoregressive-diffusion architecture claiming lossless speculative decoding via intra-model consensus, testing exact trajectory matching on 1,190 prompts across 12 domains. Under BF16, exact matching occurred in only 45% of cases for the authors' checkpoint and 43% for an independently trained model, with matching probability strongly tied to reference-model response-conditional perplexity. Despite trajectory divergence, downstream lm-eval-harness benchmarks showed no systematic degradation, while FP32 evaluation yielded exact matching on all prompts.

Hugging Face daily papers · 2d agoAI research1

Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training

NVIDIA researchers detail an end-to-end system for online draft co-training that speeds speculative decoding in large-scale long-context RL post-training.

The paper tackles scaling online draft co-training for speculative decoding in RL post-training, where rollout generation dominates cost. It extends packed, load-balanced zigzag ring attention to merge rank-local branch attention with causal main-sequence attention for context parallelism, and introduces TapChannel to transport target features across pipeline-parallel stages without changing the schedule. Experiments show co-trained drafts tracking the policy baseline with substantial rollout and end-to-end speedups up to 122B parameters and strong scaling at 256K tokens.

Hugging Face daily papers · 9d agoAI research

H Company Releases NeoMME: A Family of 260M and 800M Single-Tower Multimodal Encoders That Drop the Vision Tower and Causal Decoder

H Company released NeoMME, 260M/800M single-tower multimodal encoders matching 3.75B ColQwen2.5 on ViDoRe v3 while being 14.4x smaller, under Apache 2.0.

H Company released NeoMME, a family of 262,937,906- and 793,715,032-parameter bidirectional encoders that process text and raw 32x32 image patches in a single tower, pretrained via masked diffusion and released under Apache 2.0 with day-zero Hugging Face Transformers support. NeoMME-Retriever-260M reaches 0.523 nDCG@10 on ViDoRe v3, matching 3.75B-parameter ColQwen2.5 while being 14.4x smaller; the 800M model scores 0.556. Hierarchical token pooling with int8 and binary quantization shrinks late-interaction indexes from roughly 1.5 MB to 6 kB per page while retaining 95.19% of nDCG@10; text-only BEIR retrieval remains a weak spot.

MarkTechPost · 9d agoAI research

Why the current tech backlash feels different

The Verge's Decoder mailbag discusses the current tech backlash, arguing AI hype overstates verifiability outside software engineering.

Nilay Patel's Decoder mailbag episode addresses listener feedback on the widely discussed 'software brain' essay. He argues AI hype is concentrated on software because code is verifiable through compilation, while domains like drug discovery, math and science lack equivalent verifiability. The episode also touches on AI backlash, surveillance, data centers and upcoming midterm coverage.

The Verge · AI · 6d agoAI industry1

How to correlate Kubernetes audit logs with container runtime data

Elastic Security Labs shows how to join Kubernetes audit logs with Defend for Containers runtime data to investigate service account abuse and container escapes.

Elastic Security Labs demonstrates correlating Kubernetes audit logs with Defend for Containers (D4C) runtime telemetry in Elastic. In an Amazon EKS lab, a compromised workload service account performed discovery, read secrets, minted a token, created a privileged pod, and execed into it to attempt a container escape via nsenter and chroot. The escape wrappers appeared only in the decoded Kubernetes audit requestURI, not in runtime process events. The post covers join fields, prebuilt EQL sequence rules, and continues the control-plane correlation thread from the TeamPCP container attack scenario and the Hugging Face intrusion write-up.

Elastic Security Labs · 13d agoResearch1

Tables Decoded: DELTA for Structure, TARQA for Understanding

DELTA extracts tables into compact OTSL text and TARQA fine-tunes LLMs on it, beating VLM baselines on table QA.

DELTA separates physical structure recognition, logical structure recognition, and OCR to output tables in Optimised Table Structure Language (OTSL), a compact unified format encoding cell arrangements and content. It achieves TEDS-Structure scores comparable to state-of-the-art methods across FinTabNet, PubTabNet, and PubTables-1M, with robustness tested on a curated Hindi benchmark, TORQUE. TARQA, an LLM fine-tuned on OTSL sequences, gains 9.3 percentage points on WTQ TabQA and 9.2 points on FinTabNetQA TabVQA; code, models, and the benchmark are released on GitHub.

arXiv cs.AI / cs.LG / cs.CL · 1d agoAI research

RAPID: A Real-Time Defense Against Unauthorized Model Distillation for Text-to-Image Services

RAPID embeds defensive perturbations in a T2I model's shared VAE decoder to block unauthorized black-box distillation in real time.

The paper defends text-to-image services against model theft via black-box output-based distillation, where adversaries collect prompt-image pairs to train substitute models. RAPID integrates defensive perturbations into the shared VAE decoder using self-referenced latent maximization plus reconstruction-guided color regularization, avoiding costly sample-wise online optimization. Across four T2I models and four datasets versus five baselines, it consistently degrades substitute-model generation quality while preserving visual fidelity.

arXiv cs.CR · 2d agoAI safety & security

SpecGuard: Inference-Time Backdoor Detection For Free

SpecGuard detects backdoored LLM behavior at inference time using speculative decoding acceptance rates, adding no extra model computation.

Researchers propose SpecGuard, an inference-time backdoor detector that repurposes draft-token acceptance rates from speculative decoding as a detection signal at zero added model-computation cost. When a trigger shifts the target model toward attacker-controlled behavior, the clean draft model's acceptance rate changes, exposing the backdoor; the paper formalizes when this signal appears and shows suppressing it weakens the backdoor. Experiments across diverse backdoor types and model families show reliable detection, including stealthy cases invisible to input-level filters. Speculative decoding is positioned as a free, always-on monitor for frequently updated deployed models.

arXiv cs.CR · 6d agoAI safety & security 2 sources2

BreezeBlue/Breeze-TTS-2 — new model trending #19 on Hugging Face

BreezeBlue open-weights Breeze TTS 2, a bilingual text-to-speech model it ranks #1 among open-weight models on the Artificial Analysis TTS leaderboard.

BreezeBlue released open weights and Apache 2.0-licensed PyTorch inference code for Breeze TTS 2 on 2026-08-25. The text-to-speech model supports English and Chinese, voice cloning, reference-free voice design, voice direction, and inline vocal events like (laugh) and (sigh). Reported performance includes #1 open-weight ranking on the Artificial Analysis Elo leaderboard, under 40 ms time-to-first-audio, a 0.32 real-time factor on an NVIDIA H100, and about 7.7 GiB GPU memory for eager inference.

Hugging Face trending models · 22d agoModel release

AI models' written reasoning steps correspond to distinct internal patterns, a new study finds

KAIST and Naver AI Lab researchers show LLM reasoning steps like extraction and computation map to distinct activation patterns, strongest in middle layers.

Researchers at KAIST and Naver AI Lab defined eight recurring reasoning operations, including extraction, decomposition, formula recall, deduction, and computation, and showed they correspond to separable activation patterns in Qwen2.5-7B, Qwen3-8B, and Gemma4-31B on math tasks, with GPT-5 labeling solution segments. The separation peaks in middle layers, holds even when a computation step produces a wrong answer, and goes beyond surface-level token choice. Findings replicated on Llama-3-8B, and classifiers trained on Qwen3-8B transferred to GPQA-Diamond and MATH-500. The authors note that using internal states for error detection or mid-generation steering remains future work.

The Decoder · 4d agoAI research2

Encoded Early, Used Late: Where Transformers Begin to Act on an Inferred Partner's Expertise

Probing finds transformers represent an inferred dialogue partner's expertise in early layers long before it causally influences output.

Using ExpertCollab, a corpus of multi-turn research-planning dialogues between model-played personas at four expertise levels, researchers show that a partner's inferred expertise is most decodable in early transformer layers and decays to near chance before the network's midpoint. Counterfactual patching reveals that injecting the expertise difference at peak decodability barely changes a fixed late-layer readout, while injection past the midpoint propagates almost completely. The result bounds where readout or steering of partner-conditioned behavior must intervene, demonstrated on a single model with a synthetic corpus.

Hugging Face daily papers · 9d agoAI research

[AINews] Andrew Ng gets into AI Engineering

Andrew Ng relaunches DeepLearning.AI around AI Engineering, defining four core skills from an analysis of 10,000+ job postings and expert interviews.

Andrew Ng, cofounder of Google Brain and Coursera, relaunched DeepLearning.AI with a focus on AI Engineering, basing the curriculum direction on an analysis of over 10,000 job postings plus interviews and surveys. He identifies four key skills: building and deploying AI applications, software engineering fundamentals, effective use of coding agents, and shaping the build with product sense. The Latent Space AI News issue also recaps agent ecosystem developments, including NVIDIA's 'Skill Lift' evaluation proposal showing skill scan scores correlate only weakly (Spearman rho = 0.14) with judged quality, and Konwinski's open-source persistent-agent 'microharness' Headlong, which achieved an unattended self-debugging repair in 48 minutes.

Latent Space · 22d agoAI industry1

Large Language Models Develop Belief State Geometry In-Context

Probing six open-source LLMs on HMM-generated data shows belief states are linearly decodable (R² 0.83–0.99), suggesting in-context learning approximates Bayesian prediction.

Researchers prompted six open-source LLMs with data from 40 hidden Markov models selected for non-trivial belief structure and probed residual-stream activations for belief states (posteriors over hidden states). Belief states were linearly decodable with peak R² values of 0.83–0.99 across HMM/LLM combinations, spanning early to late layers. Patching and steering the probe-identified subspace preserved downstream prediction quality while control interventions degraded performance substantially, establishing functional relevance. The results provide representation-level evidence that in-context learning approximates optimal Bayesian prediction over a context-inferred generative model.

arXiv cs.AI / cs.LG / cs.CL · 1d agoAI research

LynnReal-Omni: Native multi-modal Video Generation for Agentic Visual Workflows

LynnReal-Omni unifies controllable video generation tasks in a 32B multimodal diffusion transformer, with a 27B Flash variant rendering 540p clips in 377 ms.

LynnReal-Omni is a native multimodal video generation framework built on a 32B shared multimodal diffusion transformer unifying text-to-video, image-conditioned generation, reference guidance, structural control, editing, restoration and long-video generation, accepting heterogeneous inputs like 3D renders and game recordings for agentic visual workflows. A dedicated 27B Flash model enables real-time rendering, producing a 22-frame 540p video in 377 ms on one H100 versus 843 ms for the full model. The work introduces a curated multi-shot audiovisual data pipeline and MSAVP, a 100-prompt, 20-metric evaluation design covering instruction following, plausibility, visual quality, temporal behavior and audio coordination.

Hugging Face daily papers · 2d agoAI research

Rethinking Heterogeneous System Disaggregation for Subquadratic Attention

SQD disaggregates LLM inference by quadratic versus subquadratic attention layers, improving energy efficiency up to 56% on heterogeneous systems versus GPU-only baselines.

SQD (SubQuadratic Disaggregation) splits decode not by operator type but by quadratic versus subquadratic attention, matching their distinct arithmetic intensity and memory footprints. For sparse attention LLMs it separates top-k selection (requiring full KV indexing) from top-k attention plus FFN; for linear and sliding-window models it separates dense attention layers from subquadratic layers plus FFN. On an adjusted 8xB200 heterogeneous proxy, tokens-per-joule improves 53% on GLM 5.2, 31% on Nemotron 3 Ultra, and 56% on Gemma 4 31B. A Rubin plus LPX analytical model shows 1.2x-1.5x tighter achievable latencies and up to 3.6x higher throughput versus attention-FFN disaggregation.

arXiv cs.AI / cs.LG / cs.CL · 4d agoAI research

Agnes-AI/Agnes-3.0-Flash — new model trending #30 on Hugging Face

Agnes AI releases open-weight Agnes-3.0-Flash Preview, a 33B multimodal model with 262k-token context under Apache 2.0.

Agnes AI released Agnes-3.0-Flash Preview, an open-weights multimodal checkpoint with 33B parameters and a 262,144-token context window under Apache 2.0. The model supports text, image, and video understanding, tool calling, and adjustable reasoning effort. The repo clarifies this preview checkpoint is distinct from the production/API Agnes 3.0 Flash model, which uses a different configuration with a 1M-token context window. Reported reference results include IFBench 74.20 and SciCode 38.08 against peers such as Qwen3.6-35B-A3B, Kimi K2.5, and MiniMax M3.

Hugging Face trending models · 5d agoModel release

Differential Privacy Meets Fixed Parameter Tractability: Algorithms and Lower Bounds

Theory paper combines differential privacy with fixed-parameter tractable encoders, improving approximation guarantees for combinatorial optimization and proving new lower bounds.

The paper studies combinatorial optimization under epsilon-differential privacy within the implicit encoder-decoder framework of Gupta et al. (SODA 2010), generalizing it to allow fixed-parameter tractable encoders. This circumvents approximation barriers inherent to polynomial-time algorithms and yields improved guarantees for fundamental combinatorial optimization problems. The authors establish the first representation-independent lower bounds: assuming a non-uniform variant of the Gap Exponential Time Hypothesis, no epsilon-DP encoder-decoder pair can achieve certain approximation guarantees with a subexponential-time decoder for sufficiently small epsilon. Representation-dependent lower bounds are also provided for larger epsilon.

arXiv cs.CR · 5d agoResearch

Meet Redis LangCache: A Managed Semantic Cache That Cuts LLM API Costs by Up to 90% and Returns Cache Hits Up to 15x Faster

Redis launches LangCache, a managed semantic cache matching LLM prompts by meaning, cutting API costs up to 90% and returning hits up to 15x faster.

Redis LangCache is a fully managed semantic caching service in public preview on Redis Cloud, accessed via a REST API with Python and JavaScript SDKs. It embeds incoming prompts, vector-searches stored entries, and returns a cached response when similarity clears a configured threshold, skipping the LLM call entirely. Redis claims up to 90% cost savings and up to 15x faster cache hits; a demo run showed 0.37 seconds versus 2.232 seconds direct inference (about 6x) with zero LLM tokens. Customer Mangoes.ai reports a 70% hit rate, 70% lower LLM spend, and 4x faster responses on a patient-care voice app.

MarkTechPost · 5d agoAI tools & infra

Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model

Dynin-Robotics unifies action, goal, and dynamics prediction in one omnimodal masked-diffusion VLA model, reaching 78.4% success on Franka Research 3 manipulation tasks.

Built on the Dynin-Omni masked-diffusion backbone, the model represents language, observations, goals, and actions as discrete tokens and is continually pretrained on roughly 1.33 million trajectories from 48 Open X-Embodiment datasets. The shared trajectory interface enables test-time scaling via goal prediction, action-candidate evaluation, and joint action/future-state refinement. It achieves competitive results on LIBERO and zero-shot LIBERO-Plus, 78.4% average success across four Franka Research 3 conditions, and up to 29.2x faster model-side action decoding from a block-parallel implementation.

Hugging Face daily papersupdated · 5d agofirst · 5d agoAI research 2 sources

Why Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs

A survey catalogs inference-efficiency techniques for video and audiovisual LLMs, mapping bottlenecks in sampling, encoding, token reduction, and LLM decoding.

This survey covers inference-efficiency mechanisms for visual and audiovisual video LLMs, reporting reductions in parameters, FLOPs, latency, memory, and token counts. It organizes methods by pipeline stage, covering frame sampling, modality encoding, connector-level token reduction, and LLM prefilling and decoding for systems built since late 2022. The authors compile accuracy-cost comparisons under shared host models and input protocols, identify gaps in audiovisual efficiency and standardized evaluation, and maintain a public repository.

Hugging Face daily papers · 7d agoAI research

Edge0/Edge0-35B-A3B-preview — new model trending #30 on Hugging Face

Edge0 released a 35B sparse MoE model running in under 3 GiB of memory at 15 tok/s via SSD expert offload and int4 quantization.

Edge0-35b-a3b-preview is a 35B-parameter MoE (256 experts, 4 active per token) built on Qwen3.5-MoE 35B-A3B, shipped as a 4-bit checkpoint with LoRA and prerouter adapters under Apache 2.0. The edge0 framework streams expert weights from SSD on demand, bounding peak active memory at 2.9 GiB and achieving 14.9-17.7 tok/s decode on a Mac mini M4 Pro (MLX backend). Recover-LoRA distillation keeps the int4 model within 3.9 points of its fp16 base (79.2 vs 83.2 average on OpenCompass benchmarks including AIME 2026, HumanEval, GPQA-Diamond, MMLU-Pro, and IFBench).

Hugging Face trending models · 8d agoModel release

ukisai/Swift-Qwen3.8-27b — new model trending #30 on Hugging Face

UkisAI releases Swift-Qwen3.8-27B, a Qwen3.8-27B derivative using 58.3% fewer thinking tokens with <1% performance loss and ~1.95x speed-up.

UkisAI released Swift-Qwen3.8-27B, a reasoning-efficient derivative of Qwen3.8-27B that cuts thinking-token usage by 58.3% while staying within 1% of base performance, yielding a 1.95x speed-up on several tasks. The model was fine-tuned by penalizing reasoning-marker tokens that trigger overthinking, plus a transfer component from BottleCap AI's ThinkingCap-Qwen3.6-27B. Benchmarks include GPQA-Diamond 88.28% (base 88.38%), MMLU-Pro 84.95% (base 85.47%), and AIME 2026 94.00% (base 98.67%), with mean-token reductions of roughly 27-46% across tests. GGUF weights are available on Hugging Face alongside enterprise licensing options.

Hugging Face trending models · 8d agoModel release

Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction

Grouped Value Attention stores grouped values and reconstructs content keys via a learned linear map, cutting KV-cache size about 45-47% versus GQA.

GVA stores only grouped values and reconstructs content keys with a learned linear map absorbed into the query at decode time, while a small shared decoupled RoPE channel preserves positional information via a separately cached positional key. At 350M parameters trained on 30B FineWeb-Edu tokens, the 16-dimensional positional variant scores 44.18 average accuracy across five tasks versus 44.36 for GQA and 43.88 for MLA. Custom decoding kernels are in development with an open-source release planned.

Hugging Face daily papers · 8d agoAI research

[AINews] OpenAI shuts off Cursor

OpenAI cut off API access to coding tool Cursor after its SpaceX acquisition, citing contract violations by Elon Musk's companies.

OpenAI disabled Cursor's access following the closing of Cursor's acquisition by SpaceX, citing its experience with Elon Musk's companies violating contracts; Cursor responded that OpenAI accounts for only 5% of its traffic. The weekly digest also covers major open-weight releases: Z.ai's GLM-5.3 (744B total/40B active, 1M context) and Tencent's Hy4-preview (770B/49B, ~#5 on Code Arena WebDev), plus Alibaba's Qwen3.8-Flash (125B/6B). vLLM published benchmarks showing no universal winner among speculative decoding methods across model families.

Latent Space · 18d agoAI industry

NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut

NVIDIA's Vera Rubin NVL72 debuts in MLPerf Inference v6.1 with up to 3.7x higher throughput than GB300 NVL72 and 99% scaling efficiency at 288 GPUs.

In its first MLPerf Inference preview submission, NVIDIA's Vera Rubin NVL72 achieved up to 3.7x higher throughput than GB300 NVL72 on Qwen3-VL and 2.5x on DeepSeek-R1. A 288-GPU GB300 NVL72 submission across four racks reached 99% scaling efficiency on the DeepSeek-R1 offline benchmark. Software optimizations delivered up to 1.6x gains over v6.0, leveraging TensorRT-LLM, vLLM, Dynamo, disaggregated serving, and NVFP4 precision.

NVIDIA Blog · 2h agoAI industry 2 sources

12 Best CNAPP Platforms Compared (2026): Features & Pricing

Independent comparison of 12 CNAPP platforms finds identical estates draw quotes 2-3x apart; Microsoft Defender for Cloud is the only fully published per-resource option.

A vendor-independent buyer's guide compares twelve CNAPP platforms including Prisma Cloud, CrowdStrike Falcon Cloud Security, Wiz, Uptycs, Aqua, Zscaler, and Microsoft Defender for Cloud on pricing mechanics, procurement leverage, and capability-per-dollar. It finds quotes swing 2-3x on identical estates because vendors define 'workload' differently. Microsoft Defender for Cloud is highlighted as the only major with fully published per-resource rates.

GBHackers · 1d agoIndustry1

Why don't machine learning research agents overfit?

Amazon researchers explain why ML research agents avoid benchmark overfitting, attributing generalization to compressibility of successful strategies.

Amazon Science summarizes the paper "What fits (into few tokens) doesn't overfit: Compression and generalization in ML research agents," which investigates why benchmark hill-climbing loops, whether run by human communities or LLM research agents, do not produce rampant overfitting. The explanation formalizes Occam's razor via a counting argument: successful ML strategies are highly compressible, so short descriptions lack room to memorize benchmark data and must capture real structure. LLM-based agents, being resettable and controllable, allow this hypothesis to be tested empirically.

Hacker News · security · 2d agoAI researchHN 21↑ · 3 comments

Retrospectively Reverse-Engineering Apple's Neural Engine

A developer reverse-engineers Apple's M1 Neural Engine architecture, mapping compute cores, MAC datapaths, and schedulers to explain the NPU's decline as transformers displaced CNN workloads.

A developer who previously maintained a reverse-engineered Linux driver for Apple's Neural Engine (ANE) published a retrospective deep dive mapping the M1 ANE's full internal architecture: compute, datapath, scheduler, memory, and execution model. The M1 ANE has 16 compute cores with 128 FP16 (or 256 INT8) MAC lanes each, totaling 2048 parallel MAC lanes, using 32-bit Q16.16 fixed-point accumulation with FP16 readout and an accumulator that saturates at 2^15. The author argues the ANE's dataflow was architected around the predictable reuse patterns of 2017-era CNN workloads (dating to the A11 Bionic), which autoregressive transformer decode broke, limiting its usefulness for general ML. With Apple's M5 folding ANE cores into GPU cores to tout LLM performance, the post frames this as the beginning of the end for the standalone NPU.

Hacker News · AI · 4d agoAI researchHN 54↑ · 7 comments2

Show HN: MultiMatte, a Promptable Image Background Removal Model

Feyn releases MultiMatte, a promptable background-removal model fine-tuned from Meta's SAM 3 via LoRA, outputting alpha mattes that beat SAM 3 on segmentation benchmarks.

Feyn introduced MultiMatte, a promptable image background-removal model built on Meta's SAM 3 (860M parameters). It modifies only 19.49M parameters (2.27%) using a rank-16 LoRA adapter and replaces binary masks with alpha mattes to handle fuzzy boundaries like hair. On the DIS-VD benchmark it scores 0.901 S-measure versus SAM 3's 0.667, and it improves on SAM 3 across all twelve evaluated splits. Training used 19,953 images for 14,000 steps with focal and Dice loss, and the merged weights are available via the nobg library and a web demo.

Hacker News · security · 6d agoModel releaseHN 21↑ · 4 comments1

Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra

Cognition released SWE-2, a coding model post-trained from Kimi K3 that scores 50.0% on FrontierCode 1.1 Main, near Fable 5.1 at 64% lower cost.

Cognition introduced SWE-2, its most advanced coding model, post-trained from the 2.8T-parameter Kimi K3 base model. It achieves 50.0% on FrontierCode 1.1 Main, 73.0% on DeepSWE 1.1, and 92.8% on Terminal-Bench 2.1, beating Grok 4.6 and SWE-1.7 while matching Fable 5.1 and GPT-5.6 Sol at a fraction of the price. The company says it scaled reinforcement learning to the multi-trillion-parameter regime for the first time, using Pareto-informed cost penalties that train all reasoning-effort levels in a single run, tripled RL environments, and NVFP4/FP8 quantization-aware training. SWE-2 is available today in Devin Desktop and CLI, with rollout on Devin Web and Fusion.

Hacker News · AIupdated · 3d agofirst · 6d agoModel release 10 sourcesHN 58↑ · 15 comments1