35 resultados para "RAG"
RAG do Zero: Busca + Geração
Dados & EmbeddingsRAG explicado passo a passo: indexar documentos com chunking e embeddings, recuperar trechos relevantes, montar o prompt e gerar a resposta.
LLMs: Como Funcionam os Modelos de Linguagem
LLMs & TextoEntenda de uma vez como funcionam os LLMs: a arquitetura transformer, o treinamento, por que eles alucinam, fine-tuning, RAG, quantização e agentes.
Agentes de IA: O Que São e Como Pensam
LLMs & TextoAgentes de IA explicados: como um LLM deixa de só responder e passa a usar ferramentas, planejar e agir — e por que isso é mais frágil do que parece.
Fine-tuning vs. RAG: Quando Usar Cada Um
LLMs & TextoFine-tuning ou RAG? Entenda a diferença, os custos e como decidir entre ajustar um modelo ou conectá-lo à sua base de conhecimento.
Embeddings & RAG: A Memória das IAs
Dados & EmbeddingsEntenda embeddings, busca semântica, bancos vetoriais e RAG — como dar memória e fontes a um modelo de linguagem, e por que tudo depende dos dados.
Unicorn, pelican, Middle-earth: OpenAI co-founder Karpathy is looking for the next AI vibe test
Dados & EmbeddingsOne paragraph of "Lord of the Rings" in, 5,500 lines of code out. Andrej Karpathy had Claude Opus 5 turn Tolkien's opening into a 3D browser scene. The article Unicorn, pelican, Middle-earth: OpenAI co-founder Karpathy is looking for the next AI vibe test appeared first on The Decoder .
Representations from Pretrained Machine-Learning Interatomic Potentials as Coarse Coordinates for Material Generation and Evaluation
Geração de ImagemarXiv:2607.28776v1 Announce Type: new Abstract: Generative machine learning is increasingly used for inorganic crystal structure generation. Most models and the corresponding evaluation approaches rely on simple forms of crystal structure representation. In this paper, we showcase the power of atom-averaged features from pretrained Machine-Learning Interatomic Potentials (MLIPs), such as MACE, for such tasks. We first introduce a distance measure that assesses the output of material generative m...
Receding-Horizon Next-Best-View Planner for Autonomous Leaf Surface Reconstruction
Dados & EmbeddingsarXiv:2607.28995v1 Announce Type: new Abstract: Accurate plant leaf modeling is fundamental to downstream tasks such as plant growth monitoring, and phenotyping for yield estimation. Autonomous robotic reconstruction for large-scale field deployment must address limitations on robot planning budget and computation resources while optimizing viewpoint utility for leaf surface reconstruction. Existing approaches either focus on rigid objects, point-cloud coverage or plant reconstruction without fu...
Event-Based Upper-Body Humanoid Teleoperation Under Challenging Illumination
Dados & EmbeddingsarXiv:2607.29227v1 Announce Type: new Abstract: We present a real-time upper-body human-to-humanoid motion imitation framework driven by neuromorphic event-based vision. This work addresses practical perceptual bottlenecks of conventional frame-based RGB sensors, specifically their difficulty in high dynamic range (HDR) scenes and rapid motions due to fixed integration times. By leveraging the Prophesee EVK4 event camera, which operates asynchronously with high temporal resolution and a dynamic ...
Tokenizer-Agnostic Engram Module
LLMs & TextoarXiv:2607.29065v1 Announce Type: new Abstract: Deepseek's Engram, a conditional memory module, was introduced to trade-off storage versus reasoning in large language models. However, the module relies on token-level $N$-gram hashing for Engram embedding lookup, introducing a tight coupling to the tokenizer used: a model with a different tokenizer would have to train its own Engram embeddings from scratch. To improve the reusability of Engram embeddings, we propose a change to the hashing routin...
TAGTorch: A PyTorch Library for Geometry, Topology, and Symmetry-Aware Machine Learning
Dados & EmbeddingsarXiv:2607.28755v1 Announce Type: new Abstract: Over the last decade, neural networks have been applied to an increasingly diverse range of applications, including data with rich geometric, topological, or symmetry-related structure. As a result, researchers have increasingly drawn inspiration from topology, algebra, and geometry. Despite this rich algorithmic development, the supporting software ecosystem remains fragmented. Many important methods exist only as research prototypes in unmaintain...
RayViT: Ray-Conditioned Visual Representations for Viewpoint-Robust Imitation Learning
Visão ComputacionalarXiv:2607.29622v1 Announce Type: new Abstract: Visual imitation learning enables robots to acquire visuomotor skills directly from images, yet RGB observations lack explicit geometric cues, making learned policies brittle to camera perturbations. To address this, we propose \textbf{Ray-conditioned Vision Transformer Encoder (RayViT)}, a lightweight architecture that injects camera geometry into pretrained ViT backbones. RayViT represents camera geometry as a Pl\"ucker ray map, patchifies it int...
Learning from Adversity: Semantic-Aware Mask Refinement through Adversarial Perturbation
Visão ComputacionalarXiv:2607.29059v1 Announce Type: new Abstract: Despite significant advances in image segmentation, even state-of-the-art models produce masks with imperfect boundaries, semantic inconsistencies, and structural errors. Mask refinement addresses these limitations, yet current approaches rely on simplistic synthetic noise that fails to capture the complex error patterns of real segmentation models. We introduce Phoenix, a novel framework that leverages adversarial learning to generate semantically...
Mirror Learning
Dados & EmbeddingsarXiv:2607.28737v1 Announce Type: new Abstract: We investigate imitation learning through the lens of third-person observation and propose a framework for mirror learning: acquiring actionable policies from passive observation. While behavior cloning (BC) excels under dense, well-aligned first-person data, it fundamentally fails to leverage the rich observational signals arising from third-person demonstrations that humans and animals routinely exploit. We introduce a method that composes (i) a ...
Fragility of Value under Imperfect Alignment
LLMs & TextoarXiv:2607.28881v1 Announce Type: new Abstract: As more responsibility is placed upon AI systems, it becomes increasingly important to guarantee that these systems are aligned with humanity. A common fear in AI safety is that human value is fragile -- that is, optimizing too heavily for an imperfect proxy to human values will lead to a catastrophic outcome. In this paper, we present a model of the alignment problem where an agent undergoes idealized alignment training that guarantees its value f...
LAWFUL: Law-Aligned Witness for Faithful Use of Latents
Dados & EmbeddingsarXiv:2607.28672v1 Announce Type: new Abstract: When a neural network predicts a physical system accurately, has it learned the governing law as formal, structured knowledge, and if so, does the network's internal computation actually use that representation throughout the law's domain of validity? We identify four interpretability gaps that limit answering these questions for {\em physics laws over continuous variables}: the absence of a coverage-aware causal-consistency measure over continuous...
Auto-JEPA: A Latent World Model of Continuous Intent for End-to-End Autonomous Driving
Robótica & RLarXiv:2607.29031v1 Announce Type: new Abstract: Existing autonomous-driving world models typically perform dense prediction of future videos, occupancy states, BEV representations, or agent motion. We argue that planning need not reconstruct the complete future world, but only focus on scene features that affect future ego action. Based on this perspective, we propose Auto-JEPA, an action-oriented latent world model that learns continuous future driving intent through joint-embedding prediction....
Mitigating Class-Tail Undercoverage in Medical Vision-Language Models under Clinical Shift
MultimodalarXiv:2607.28696v1 Announce Type: new Abstract: Medical vision-language models (VLMs) can retain high observed marginal coverage after clinical shift while substantially under-covering an individual disease class. The affected class varies with acquisition protocol and backbone geometry, so source prevalence does not reliably reveal the failure. Existing localized and tail-aware conformal methods respectively adapt to test neighborhoods and source-frequency tails, leaving held-out class-wise cov...
FibVLA: An Efficient Temporal Vision-Language-Action Model with Fibonacci Sampling
MultimodalarXiv:2607.29596v1 Announce Type: new Abstract: Vision-language-action models (VLAs), which leverage the cognition of multimodal information to infer physical-world actions, provide a generalized solution for embodied AI applications. Conventional VLAs usually concentrate on current digital cognition. While some efforts are made to enhance VLAs' reasoning capabilities by capturing temporal information, encoding the long-context history causes an efficiency-decreasing issue. To reconcile the conf...
Parameter-Efficient Fine-Tuning for Spiking Point Cloud Models
LLMs & TextoarXiv:2607.29048v1 Announce Type: new Abstract: Spiking Neural Networks (SNNs) offer energy-efficient solutions for point cloud analysis on resource-constrained devices through event-driven computation. However, existing pre-trained spiking point cloud models rely on full fine-tuning for downstream task adaptation, incurring substantial parameter and storage overhead. Furthermore, binary spike propagation suppresses task-relevant sub-threshold information. To address these issues, we propose Spi...
Flow Matching with Missing Data
Dados & EmbeddingsarXiv:2607.28698v1 Announce Type: new Abstract: Flow matching assumes fully observed training data, which many real-world applications rarely provide. We propose Missing-Data Flow Matching, which treats the missing coordinates of training samples as latent variables and averages the flow matching loss over the values they could take. We first prove the correction is exact rather than approximate. Under missing completely at random with true completions, the incomplete-data objective equals the c...
BLADE: Boundary-Expanded and Layer-Adaptive Dynamic Exit for Efficient LLM Reasoning
LLMs & TextoarXiv:2607.28966v1 Announce Type: new Abstract: Large language models often improve task performance by generating long reasoning traces, but the resulting computation is frequently wasted on redundant verification and revision. Existing probe-based early-exit approaches mainly inspect explicit self-doubt expressions, leaving many earlier termination opportunities undetected. Expanding inspection to ordinary reasoning boundaries improves coverage, but also exposes highly diverse intermediate sta...
M3-DuplexBench: A Multi-Turn, Multilingual, Multidomain Benchmark for Full-Duplex Spoken Dialogue Models
Dados & EmbeddingsarXiv:2607.29125v1 Announce Type: new Abstract: Full-duplex spoken dialogue systems (FDSDSs) can listen while speaking, enabling natural behaviors such as smooth turn-taking, backchannel handling, and user barge-in handling. However, fair comparisons in multi-turn conversations remain a challenge. In addition, existing benchmarks provide limited coverage of languages and dialogue domains. We propose M3-DuplexBench, a multi-turn, multilingual, multidomain benchmark for FDSDSs. M3-DuplexBench supp...
ViSAGE: Constructing Self-Correcting Memories for Long-Form Video Understanding
LLMs & TextoarXiv:2607.28678v1 Announce Type: new Abstract: Multimodal agents operating in long-horizon environments must build and continually update multimedia memories to support entity-consistent, temporally grounded reasoning. However, existing agentic memory approaches often discard fine-grained dentity cues under aggressive compression and segment-wise processing. They also rely heavily on vector similarity retrieval, which can surface semantically related yet identity-mismatched evidence, leading to...
DeepSeek Upgrades DeepSeek-V4-Flash-0731 with Major Agentic and Coding Gains
LLMs & TextoDeepSeek published DeepSeek-V4-Flash-0731 on Hugging Face and moved the official V4-Flash API into public beta on July 31, 2026. The model card is explicit that this is the official release superseding the preview, and that the architecture and size are unchanged. The gains come from re-post-training, not a new design. The checkpoint ships with the […] The post DeepSeek Upgrades DeepSeek-V4-Flash-0731 with Major Agentic and Coding Gains appeared first on MarkTechPost .
Aschenbrenner's AI thesis could be correct, his timing and leverage were not
Dados & EmbeddingsLeopold Aschenbrenner's AI hedge fund Situational Awareness had to unload nearly its entire publicly traded portfolio to Ken Griffin's Citadel after racking up heavy losses on leveraged AI stock positions. Just days earlier, Aschenbrenner had reported a six-month return of 439 percent and pulled in fresh capital. Then margin calls forced the fire sale. The article Aschenbrenner's AI thesis could be correct, his timing and leverage were not appeared first on The Decoder .
Stack v3: 114 TB de código aberto e a pergunta de quem foi consultado
Dados & EmbeddingsO maior conjunto de dados de código-fonte já publicado abertamente: 224 milhões de repositórios, 770 linguagens, cerca de 5 trilhões de tokens depois da filtragem. É quase dez vezes o Stack v2 — e o ponto não é o tamanho, é que agora qualquer laboratório pode treinar com a mesma base que os grandes.
Corrigible Assistance in One Round: Pragmatic-Pedagogic Best Response
LLMs & TextoarXiv:2607.27508v1 Announce Type: new Abstract: Assistance games formalize human-robot collaboration under asymmetric information: the human knows the goal, while the robot must infer it from observation and interaction in order to assist effectively. In general, computing optimal assistance game strategies online is intractable, since exact solutions require planning in a POMDP. We identify a class of assistance games in which pragmatic-pedagogic reasoning resolves goal uncertainty in a single ...
LayerRAG-Bench: A Cross-Layer Reliability Benchmark for Agentic Retrieval-Augmented Generation
Dados & EmbeddingsarXiv:2607.27353v1 Announce Type: new Abstract: Agentic retrieval-augmented generation systems can produce answers that appear grounded while failing at the evidence, tool-contract, authorization, or session-state layer. We introduce LayerRAG-Bench, a controlled cross-layer reliability benchmark with 8 enterprise domains, 240 tasks, 9 fault scenarios, 2 contract modes, and 38,880 live task-level records across nine models from OpenAI, Anthropic, and Gemini. Schema normalization raises schema-dri...
HSS-Synth: Humanities and Social Sciences Data Synthesis for LLMs
LLMs & TextoarXiv:2607.27379v1 Announce Type: new Abstract: High-quality, diverse data are vital for large language models (LLMs) but remain scarce and costly. Data synthesis is a viable alternative and succeeds on closed tasks, yet the humanities and social sciences (HSS) are overlooked, and their open-ended nature makes synthesis challenging. Moving beyond prior capability-centric, fragmented attempts, we adopt a subject-centric paradigm, define the first HSS domain system covering 14 mainstream fields, a...
GoGoTB: Agentic RTL Verification with Specification-Grounded Coverage Closure
LLMs & TextoarXiv:2607.26181v1 Announce Type: new Abstract: Functional verification dominates integrated circuit (IC) front-end engineering effort, and a single missed bug that escapes to silicon can trigger a costly respin. Recent large language models (LLMs) offer new opportunities to automate this process, yet existing LLM-based approaches generate each component through independent single-turn calls with no shared context, leaving interface mismatches undetected and reported coverage disconnected from s...
Cross-Embodiment Transfer via Behavior-Aligned Representations
Dados & EmbeddingsarXiv:2607.27549v1 Announce Type: new Abstract: Recent progress in large-scale imitation learning for robot manipulation has been driven by leveraging datasets across a wide range of robot embodiments. However, achieving significant cross-embodiment transfer is often still challenging. In this work, we study the role of using behavior-aligned representations (e.g., object bounding boxes, language motions, end-effector traces of robot motion) in vision-language-action (VLA) models to promote cros...
TEA-AgriVLN: Traversability Estimation Alarm for Agricultural Vision-and-Language Navigation
LLMs & TextoarXiv:2607.28474v1 Announce Type: new Abstract: Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires an agent to follow a natural language instruction, predicting a sequence of low-level actions to navigate a robot from a starting point to a target location. The A2A benchmark and the AgriVLN method pioneeringly extended VLN-CE from indoor scenes to agricultural scenes, while we observed a challenging distinction: In indoor scenes, whether a zone is traversable tends to be ...
OmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic Grounding
LLMs & TextoarXiv:2607.27155v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly expected to assist users in completing tasks. However, existing benchmarks provide limited support for evaluating whether agents can carry out office-suite workflows at a reasonable cost. We introduce OmegaUse-OfficeVal, a benchmark for evaluating LLM agents on long-horizon office-suite tasks with task-level economic grounding. The benchmark comprises 100 tasks derived from office-suite requests pr...
Models for minimalist RAG: B1ade 335M Embedding and 1B Parameter Small Language Models
Dados & EmbeddingsarXiv:2607.27506v1 Announce Type: new Abstract: Language and embedding models used in RAG systems are conventionally assumed to require large-scale pretraining and explicit grounding supervision. We present B1ade, an efficient RAG architecture comprising two purpose-built components: a compact embedding model and a purpose-built SLM. B1ade-embed, a 335M parameter retrieval model constructed via parameter-free fusion of five pretrained encoders achieves top MTEB scores among sub-500M models with ...