Loading…
Topic
617 posts about Machine Learning. Every summary links to the original.
See the idea
A walkthrough of how a 3D CT model turns a voxel volume into a structured diagnostic report: isotropic resampling, 3D patch tokens, spatial encoding, and language-model reasoning.
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Newest first
AI applications are presented as pipelines that chain image, text, audio, and dataset operations, with intermediate values otherwise investigated through print-debugging. gr.Workflow turns these pipelines into graphs of typed nodes, served as a drag-and-drop canvas where each node runs independently and its result remains visible. Operators can be Python functions, Hugging Face Inference Providers models, Gradio Spaces, Hub dataset rows, or GPU-backed functions using ZeroGPU, while fan-out connections run multiple analyses or generations in parallel. The same workflow also exposes labeled outputs as REST endpoints callable through the Gradio client or curl, and can be deployed to Hugging Face Spaces with one command.
Existing multimodal models struggle with structural fragmentation, often sacrificing visual fidelity via discrete tokenization or breaking structural symmetry by pairing causal text generation with diffusion-based denoising. STARFlow2 addresses these limitations by leveraging autoregressive normalizing flows, which share the causal mask, key-value cache mechanism, and left-to-right structure of large language models. Built upon the Pretzel architecture, the system vertically interleaves a frozen pretrained vision-language model stream with a TARFlow stream through residual skip connections under a single causal mask. A deep-shallow flow design combined with a unified FAE latent space allows both text and continuous visual outputs to enter the key-value cache directly without re-encoding. Benchmark experiments validate strong performance across multimodal understanding and continuous image generation tasks.
Salesforce addressed an operational challenge where sellers faced roughly 12,000 dashboards and over 20 applications outputting machine learning predictions without clear next steps. The engineering team reframed machine learning outputs as raw signals rather than standalone answers. To bridge the gap between assessment and action, they built a Next Best Action layer that combines model signals, business logic, and contextual institutional knowledge into actionable recommendations. They integrated this layer with an AI agent using Model Context Protocol (MCP) tool contracts, enabling dynamic discovery and explicit handling of missing data. Finally, rather than introducing a separate dashboard destination, the agent serves on-demand recommendations directly inside Slack where sellers already collaborate.
Databricks introduces updated IDE and CLI integrations that connect VS Code, Cursor, or a terminal directly to Databricks compute, workspace, and Unity Catalog. Existing Databricks Connect supported local Spark development, but remote non-Spark workloads and dependency synchronization remained pain points. The new SSH tunnel supports interactive execution and debugging of Python and SQL workloads on Serverless, AI Runtime, and dedicated clusters, while keeping files and dependencies aligned with Databricks Runtime and the workspace. CLI flags can select serverless, GPU-backed AI Runtime, or a dedicated cluster, and can start the tunnel in VS Code or Cursor; additional options configure base environments and usage policies, while the IDE can browse Unity Catalog assets.
Serving agentic AI at more than 3,000 output tokens per second while retaining a 100K-token KV cache creates a small-batch tensor-parallelism challenge: fixed communication and coordination costs can offset distributed-compute gains. NVIDIA Groq 3 LPX addresses this with compiler-scheduled execution across 256 LP30 local processing units, 128 GB of collective SRAM, and 96 112-Gbps C2C links per chip, planning transfers before execution and overlapping communication with computation at 320-byte-vector granularity. Artificial Analysis measured 3,431 output tokens per second at 100K context on Gemma 4 31B, and 3,382 tokens per second at 10K context. On SPEED-Bench coding tasks, the same system reached a median 4,767 tokens per second and a P80 of 5,520. The post describes pairing LPX with Vera Rubin NVL72 through prefill-decode, attention-FFN, or external-drafter speculative-decoding configurations for projected large-model agentic workloads.
Red Hat OpenShift AI 3.5 integrates Ray cluster runtime images with preinstalled Training Hub packages to support distributed fine-tuning algorithms including LoRA, SFT, OSFT, and GRPO. Using the CodeFlare software development kit within a Jupyter workbench, developers can provision an elastic single-GPU Ray cluster attached to shared ReadWriteMany persistent storage containing model weights and training datasets. The tutorial demonstrates fine-tuning the Qwen2.5-1.5B-Instruct model on the public sql-create-context dataset to produce direct, bare SQL output without explanatory conversational text. After training completes, the workbench loads and merges the resulting LoRA adapter checkpoints with the base model to evaluate output format correctness against sample database schema questions. The RayCluster automatically tears down resources once the job concludes, while the shared pipeline architecture allows switching between supported fine-tuning algorithms by modifying dataset inputs and configuration parameters.
Visual chain-of-thought methods allow multimodal large language models to reason through intermediate image generation, but synthesizing explicit visual steps creates major inference overhead during proactive video reasoning. To eliminate runtime synthesis, Internalized Visual Thinking introduces a post-training framework that jointly optimizes textual prediction and next-embedding prediction across unlabeled videos. By predicting latent representations of future frames alongside target text answers during training, the model captures motion, object transitions, interactions, and latent intent. At inference time, the model produces answers directly without generating or re-encoding future frames. IVT outperforms text-only post-training across six evaluation settings while matching or exceeding Visual CoT performance and cutting end-to-end latency by more than 5x.
Cloudflare announced Bot Preference Sync, a feature across all customer tiers that automatically reflects dashboard AI bot configurations directly in a site's robots.txt file. Website operators previously faced challenges synchronizing static robots.txt directives with edge enforcement rules, which led some crawlers to ignore stated preferences or attempt bypasses. Under this system, Cloudflare generates and prepends directives for Search, Agent, and Training categories to existing robots.txt files using crawler lists tracked in BotBase. Mixed-use crawlers performing both search indexing and training must meet strict transparency criteria, including honoring opt-outs and providing page-level visibility, to avoid edge blocks when training is disallowed. Additionally, onboarding workflows allow ad-supported publishers to automatically set AI model training to Disallow by default while retaining search visibility.
Quantitative strategies need instrument groupings that capture both ordinary co-movement and dependence during market stress, while remaining stable and inexpensive to recompute. The workflow constructs rolling absolute Pearson-correlation and tail pairwise dependence matrices (TPDM), factorizes them with nonnegative SymNMF using AdaptGrow, and derives soft loadings, hard argmax labels, and structural-break diagnostics. A trace-based formulation reduces estimated peak storage from approximately 20n2 to 4n2 bytes, while row-sharded distributed execution communicates O(nk) data; the system uses eigenspectrum-based selection between full-batch and block-stochastic AdaGrad. Across 250 synthetic windows, the setup tests planted membership changes and a joint tail-stress episode: ARI is intended to identify reclassification, whereas TPDM reveals co-crash dependence that ordinary correlation can miss. Reported factorization timings include 13.0 and 12.4 seconds for 100,000 instruments across four GB200 GPUs, and approximately two and four minutes for one million instruments across 64 GPUs on 16 nodes, not end-to-end timings.
amazon.science
AI agents often struggle to execute standard operating procedures because real-world business workflows contain implicit domain knowledge and procedural ambiguity that standard benchmarks overlook. To evaluate agent reliability, researchers introduced SOP-Bench, an open framework featuring more than 2,000 runnable tasks across 12 industrial domains paired with functioning tool interfaces and ground-truth answers. Domain experts authored the workflows, while an Anthropic Claude 3.5 Sonnet v2 model generated mock APIs, tool code, and edge-case datasets before expert verification. Experiments evaluating function-calling and reasoning agents across 11 frontier models revealed that newer models did not consistently outperform older versions, and providing 20 extraneous tools nearly halved task success. The benchmark, baseline agents, and evaluation code are released on GitHub and HuggingFace for reproducible testing.
NVIDIA’s Agentic Variation Operators (AVO) is a general-purpose agent architecture designed to make frontier language models reliable on extended, multistep tasks. It combines tool use and execution with persistent memory, an iterative inspect-plan-implement-evaluate loop, and a supervisor that can redirect work when progress stalls. In a seven-day GPU-kernel optimization study, AVO explored more than 500 directions, produced 40 committed versions, and outperformed cuDNN by up to 3.5% and FlashAttention-4 by up to 10.5% across evaluated configurations. On the ARC-AGI-3 public set, the same architecture achieved a 100.00 RHAE score across 25 environments, completing all 183 levels in 6,624 actions; the post cautions that comparisons with other systems are not controlled ablations.
Papers with Code needs a search engine that supports exact titles and arXiv identifiers, semantic requests, navigational queries, typos, and fast responses when model services are cold or unavailable. The system uses hybrid retrieval: PostgreSQL full-text search provides a lexical baseline, pgvector stores dense embeddings, and reciprocal rank fusion combines them. Full-corpus embeddings for more than 110,000 papers run in burstable GPU Jobs from repeatable-read database snapshots, with checksummed JSONL inputs and resumable Parquet outputs stored in Buckets. An Inference Endpoint serves low-latency query embeddings and bounded hourly updates for changed papers, while lexical retrieval remains a fallback. The embedding contract pins the model revision, dimension, prompts, normalization, and source hash; production uses Qwen/Qwen3-Embedding-0.6B with 256-dimensional L2-normalized vectors, and stored vectors also support related-paper retrieval without a request-time model call.
Public voice AI benchmark scores can overstate speech-recognition ability when models learn benchmark-specific patterns rather than improve underlying transcription. Traditional evaluations may miss reliability, naturalness, contextual appropriateness, and other real-world conditions, motivating held-out sets in Real World VoiceEQ, the Open-ASR Leaderboard, and the Far-field ASR Leaderboard. The study introduces three tests and evaluates 11 widely used open-source ASR models, finding that several reproduce VoxPopuli and LibriSpeech reference transcripts when audio contradicts them, words are silenced, or two written forms sound identical. Models also sometimes use acoustic cues to identify dataset membership and select its expected transcription policy, while fresh post-cutoff data and other interventions often restore more audio-faithful output; the findings support fully held-out evaluation and broader metrics than a single public WER score.
Grab designed Jarvis Pro as an AI assistant for account managers turning complex account data into actionable merchant conversations. Early prototypes generated polished but operationally unsound advice, such as pushing promotions when order drops were caused by outlet pauses and fulfillment issues. To address this, the team implemented a "route first, answer later" design that uses a prompted LLM classification step constrained by a routing guide to dictate context, tool paths, answer structure, and guardrails before generation. The pipeline also isolates narrow memory decisions, reconciles metric freshness across data sources, and caps tool calls to maintain usability. In offline evaluations across 351 realistic prompts, routing matched safe routes 99.4% of the time while answer quality scores in a portfolio subset improved from 78.5 to 91.0.
Technical project managers historically spend 60 to 70 percent of their working time managing routine coordination tasks, including manual spreadsheet updates and obsolete status reporting. Engineering organizations are addressing this overhead by integrating generative artificial intelligence, large language models, and agentic workflows into delivery pipelines. Automated agents ingest telemetry directly from Git commits, pull request comments, and CI/CD logs to assess system state while machine learning models execute Monte Carlo simulations to forecast delivery risks. Although this automation reduces administrative overhead below 30 percent, implementations face practical obstacles like inconsistent ticket definitions across teams and algorithmic over-reliance on uncontextualized historical patterns. Consequently, technical project managers are transitioning into strategic roles that coordinate hybrid human-agent workforces while providing critical judgment and organizational context.
Liquid AI released DSpark draft checkpoints for LFM2.5-1.2B-Instruct, LFM2.5-2.6B, and LFM2.5-8B-A1B, adding speculative decoding with a minimal memory increase. DSpark uses a DFlash-style parallel backbone conditioned on target context features, a lightweight sequential Markov-chain head, and a confidence-scheduled verifier that prunes low-confidence suffixes. The draft models are simplified attention-only networks with five layers and a block of nine, trained for 15 epochs on SFT, chat, code, and function-calling data; each contains roughly 300M parameters. Across five datasets, LFM2.5-2.6B averaged 2.67x speedup on an H100 and 2.27x on an M4 Max, while reducing multi-tool function-calling latency by 57% on average. Greedy decoding remains identical to baseline by construction, and integrations are available in llama.cpp and SGLang, with on-device gains for the 8B-A1B model limited by its current MoE Metal implementation.
Generative recommenders (GRs) recast recommendation as next-item or next-action sequence prediction conditioned on user histories, replacing traditional embedding-similarity objectives and potentially unifying retrieval and ranking. The post presents HSTU, which uses timestamp-ordered interleaved items and actions, SiLU-based attention weighting, relative attention bias, and gating, alongside Semantic IDs, which derive a smaller vocabulary through hierarchical clustering of item embeddings. For production, recsys-examples combines TorchRec, DynamicEmb, Megatron-Core, recommender-specific KV and decoding components, while nv-embedding-cache provides tiered GPU, CPU, and remote embedding storage. On an H100, the Semantic ID path was reported as 1.85x higher in online throughput and 46% lower median latency than SGLang under the stated long-context, beam-search workload, and DLRM v3 reached 99,997 queries/sec.
Cross-lingual knowledge transfer is critical for training multilingual language models when target-language data is scarce and downstream tasks demand reasoning or world knowledge. Existing transfer approaches often depend on scarce resources such as large parallel corpora, translation systems, auxiliary models, or extra training phases. To address this limitation, the authors introduce LINK, a data-level intervention method that operates directly during pretraining via bilingual vocabularies. LINK swaps randomly selected words in a portion of the high-resource English corpus with word-level translations according to a specified replacement ratio. Across evaluations spanning eight languages and five model sizes, the technique delivered notable performance gains on downstream target-language tasks while achieving up to a 2x speedup in training to reach equivalent performance.
Scaling language models demands large volumes of data, but high-value target sources like low-resource languages and specialized domains remain inherently scarce. Practitioners frequently combine limited target data with abundant generic data, creating a balance between domain underexposure and excessive repetition leading to overfitting. An empirical evaluation across more than 2,000 language-model training runs demonstrates that mixture training tolerates substantially higher data repetition than single-source training, allowing scarce target corpora to be repeated 15 to 20 times. Researchers introduced a repetition-aware mixture scaling law that accounts for both the diminishing returns of repeated target tokens and the regularizing effect of generic data. Optimizing this scaling law offers a principled framework for calculating effective mixture configurations and practical pretraining recommendations under severe data constraints.
Code-switching automatic speech recognition faces major performance challenges due to the scarcity of code-switching training data. To mitigate this data limitation, an iterative pseudo-labeling framework leverages a large unlabeled audio corpus to generate semi-supervised speech datasets. The system structure executes across three sequential phases: pseudo-label generation, two-stage bilingual model pre-training with supervised code-switching fine-tuning, and progressive iterative improvements. Successive refinement cycles further enhance the neural model's recognition accuracy when processing complex code-switching scenarios across alternating languages. When evaluated on the SEAME benchmark, this training pipeline achieved Mix Error Rate reductions of 6.35 percent on the devman subset and 8.29 percent on the devsge subset.
The best engineering reads, once a week.
