Tema

Recuperación y RAG

1217documentos

102últimos 7 días

DecisionRank: On-Policy Distillation for Decision-Model Rerankers

Decision models target structured tasks by directly predicting choices or scores, avoiding autoregressive decoding latency. However, transferring a generative LLM's ranking knowledge to a decision model remains challenging: the teacher expresses preferences through a ranked list of document ID sequences, while the student determines the ranking using scores from the first-token position. To bridge this gap, we introduce DecisionRank, a decision model trained through item-space on-policy distillation (OPD). Item-space OPD constructs the student's rollout by sorting its document scores into a ranked sequence. A generative teacher takes the resulting ranking list as inputs and provides conditional feedback to directly supervise those scores. Combined with offline ranking supervision, DecisionRank achieves state-of-the-art performance on both general and reasoning-intensive retrieval benchmarks such as TREC DL, BEIR, and BRIGHT. Across four benchmarks, item-space OPD consistently outperforms alternative training objectives. Even at 1.7B parameters, DecisionRank surpasses its 27B generative teacher on TREC DL19--23. At 8B, it outperforms its generative listwise counterpart while running 24.5faster on TREC DL and 16.7faster on BRIGHT.

arXiv AI Papershace 13 días

Subquadratic Attention with a Hierarchical Tree-Structured Indexer

The quadratic scaling of attention is a fundamental bottleneck limiting the long-context capabilities of large language models (LLMs). A variety of approaches have attempted to circumvent this, but they either fail to match attention's expressiveness (e.g., sliding-window or linear attention variants with fixed-size state) or only achieve constant-factor speedups (e.g., sparse attention methods whose indexers themselves scale quadratically). In this paper, we derive an asymptotically subquadratic sparse attention method based on a hierarchical, tree-structured indexer, which we call Zyphra Hierarchical Attention (ZHA). ZHA constructs a tree of recursively compressed key summaries over the KV cache, which future queries traverse from the top down to retrieve a sparse subset of keys. Crucially, by iteratively growing the tree's depth, the total cost including routing and final attention grows sublinearly in sequence length, resulting in overall subquadratic (but superlinear) scaling for attention. We demonstrate that ZHA can be cheaply applied to existing pretrained models with only a short KL distillation training stage over a few billion tokens, and maintains performance on long-context retrieval and challenging reasoning, mathematical, and agentic coding evaluation tasks. Moreover, we show that our method is highly robust to tree geometry including very deep trees of up to 6 levels. Finally, we develop a detailed performance model and show that our approach scales subquadratically, dramatically outperforms competing methods beyond 1 million context, and ultimately has the potential to extend effective context windows to the hundreds of millions or billions of tokens.

arXiv AI Papershace 13 días

RangeSWA: Training-Free Length Extrapolation for Continuous Diffusion Language Models

Continuous diffusion language models show promising potential in text generation tasks, but are prone to problems such as text repetition, semantic drift, and generation quality degradation when the generation length exceeds the training length. To address these challenges, we propose RangeSWA, a training-free length extrapolation method. The method combines local sliding-window attention, remote block summaries, and dynamic content retrieval to preserve local details and support long-range context access. In addition, generation degradation caused by length extrapolation is mitigated through positional remapping, conditioning attention correction, and denoising-stage-adaptive window scheduling. We conducted experiments on multiple continuous diffusion language models, and the results show that RangeSWA can effectively reduce generative perplexity and text repetition rates in long-text generation.

AWS Machine Learning Bloghace 13 días

How Worldwide Express automated freight exceptions with Amazon Bedrock

Worldwide Express automated its less-than-truckload freight billing-exception process with a multi-agent solution built on Amazon Bedrock. Six months after launch, the company cut active exceptions by 75%, improved time-to-cash by four weeks, and unlocked $10 million in free cash flow.

arXiv AI Papershace 13 días

RAMA: Retrieval-Augmented Multi-Agent Routing with Episodic Memory for LEO Satellite Networks

Low-Earth orbit (LEO) satellite networks present a challenging routing environment due to rapidly changing topology, traffic, and link conditions. Existing deep reinforcement learning (DRL) routing methods typically rely on current observations and information encoded in network weights, limiting their ability to exploit successful past experiences. We propose RAMA, a retrieval-augmented multi-agent DRL framework with Approximate Information State (AIS), in which each satellite combines its local observation, recent routing history, neighboring state information, and retrieved entries from a per-satellite episodic memory. Retrieved experiences are integrated through a Transformer-based decision network trained with AIS-based auxiliary objectives. With weights frozen in a 140-satellite Kepler constellation, RAMA reduces mean end-to-end latency by 13.4% and 99th-percentile latency by 66.7% relative to the state-of-theart double deep Q-network (DDQN) routing method, mainly by avoiding long detours. Updated online, RAMA remains stable while DDQN degrades. On the 1,584-satellite Starlink shell 1, the reduction in mean latency grows to 93.2%, and updating RAMA online lowers its mean latency by a further 25.5%. These results demonstrate the effectiveness of retrieval-augmented memory for scalable decentralized routing in LEO satellite networks.

arXiv AI Papershace 13 días

A richness-sensitive Acoustic Biodiversity Index from species-level detections: classifier benchmarking, false-positive control and validation against standard soundscape indices in an African savanna and an urban zoo

Acoustic indices are widely used as rapid proxies for biodiversity, but they summarise the distribution of sound energy rather than the species producing it and often track richness poorly. We revise and test an Acoustic Biodiversity Index (ABI) built from species-level detections. Its primary form, ABI_E, is Shannon diversity of detections normalised by the regional species pool expected at the site and season (H / ln S_ref); temporal stability T is reported as a separate component. We first benchmarked four classifiers on a strongly labelled tropical dataset (1,216 calls, 24 classes) with recording-grouped cross-validation. BirdNET embeddings + logistic regression performed best (macro-F1 0.750 +/- 0.029), and random frame-level splits inflated accuracy from 0.41 to 0.70. We then applied BirdNET v2.4 with a location and season filter to recordings from a riverine savanna in the Zambezi Valley, Zimbabwe, and from three habitats at Emirates Park Zoo and Resort, Abu Dhabi, United Arab Emirates. A stricter post-hoc location filter removed 517 high-confidence detections of 9 implausible species, all at the zoo. After diel-matched rarefaction to equal effort (6 h), the savanna site held 21 species (Hill number 1D = 9.83; ABI_E = 0.38) against 1.3-13 species (ABI_E = 0.03-0.21) at the zoo sites. Across 149 recorded hours, ABI_E correlated strongly with richness (Spearman rho = 0.93), whereas six standard indices did not (best: ACI, rho = 0.48). The stability term was lower at the savanna site (0.39) than at the zoo (0.65-0.67), showing that it penalises the natural diel cycle of bird activity. Species-level indices with explicit false-positive control, a declared regional reference pool and effort standardisation provide a more interpretable biodiversity indicator than energy-based acoustic indices.

arXiv AI Papershace 13 días

DAMI: Dual-Axis Memory Initialization for Enhanced Long-Context Retrieval in Gated Linear Attention

Linear attention scales linearly with sequence length, but its finite-state memory can limit long-context retrieval. In gated linear recurrences, memory retention depends on state decay, which is shaped jointly by gate parameters A_log and dt_bias and by input activations. In controlled Gated DeltaNet (GDN) experiments, both parameters remain highly correlated with their initial values after training, highlighting the importance of initialization. Therefore, we introduce Dual-Axis Memory Initialization (DAMI), which jointly configures these parameters before training. DAMI assigns geometric timescales across heads for varied retrieval distances and a low-high-low input-sensitivity profile across layers. A small activation sample calibrates the GDN-family implementation to match both behavioral targets without adding trainable parameters. Across GDN, KDA, GDN2, Mamba-2, and RWKV-7, DAMI improves average RULER-style retrieval accuracy by 5.29 to 16.71 percentage points over the corresponding random-initialization baselines. Performance on language modeling, commonsense reasoning, and 2K real-world retrieval remains broadly comparable to those baselines. Our code is available at https://speedai.com/research/dami.

arXiv AI Papershace 14 días

MEMIG: Self-evolving Memory Construction for Antibody Optimization

Memory has become central to self-evolving LLM agents, which store and retrieve experience from past rollouts to guide future decisions. Existing systems represent this experience as free-form natural language and maintain it by rewriting, which fits tasks whose experience is narratable. However, in scientific discovery and optimization, experience is often a quantity measured by an oracle under a specific context. Rewriting it as text leaves no index for where a record should be, replaces a measurement with a description, and cannot express how interventions interact. To address this, we argue that agent memory for biomolecule optimization should instead be indexable, quantitative, and structured, and we introduce MEMIG. This memory construction framework realizes these three properties. In detail, MEMIG addresses each insight with a domain-grounded identifier and accumulates the quantitative effect measured by a property oracle, so that updates remain local and retrieval becomes an exact lookup. Next, MEMIG connects insights whose interventions are applied together and measures their epistasis by decomposing each evaluated set. Our results show that MEMIG improves performance in most settings across in silico antibody optimization tasks.

arXiv AI Papershace 14 días

From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models

Real-world image editing requires coordinating multiple visual transformations with content preservation and reference constraints, yet aggregate scores can readily obscure unmet local requirements. Effective evaluation must therefore identify local failures and translate them into priorities for subsequent model improvement. We introduce WeGenEditBench, comprising 2,740 evaluation instances and 148 task types across single-image, text-intensive, and multi-reference editing. Pre-generated checklists shared across models decompose each request into atomic requirements; item-level deductions and visual evidence are established first and then summarized into scores for edit execution, preservation, visual quality, and task-specific correctness. To support the experience-acquisition component of recursive self-improvement (RSI), we develop WeGenEvoStudio, a unified multi-agent harness for RSI-oriented learning that integrates model-conditioned failure diagnosis, retrieval-grounded data construction, supervised adaptation, and re-evaluation. Among 15 evaluated configurations, GPT-Image-2 leads single-image editing with an Overall score of 4.076, while Seedream-5.0-Pro leads text-intensive editing at 4.291. Evolution experiments across four generation and editing benchmarks demonstrate substantial targeted gains: on their report-selected Top-5 single-image targets, FLUX.2 [dev] improves Overall from 1.083 to 3.517, while Qwen-Image-Edit-2511 gains 13.2%. Together, our benchmark and harness provide a foundation for developing more reliable image models through interpretable evaluation and feedback-driven learning.

arXiv AI Papershace 14 días

Context as Infrastructure: Knowledge Compilation for Retrieval in Agentic Systems with Harnesses

Retrieval-augmented generation is becoming agentic. A search harness writes queries, reads results, reformulates, and decides when to stop. This largely solves accuracy: a harness with only BM25/grep and MiniMax M3 answers 85% of BrowseComp-Plus and 91% on multi-hop FRAMES in our runs, without compiled or vector indexes. It exposes a context problem: an undisciplined loop repeats queries, rereads documents, and fills the window with unused passages. The missing layer is knowledge compilation: a configurable ingestion pipeline where LLMs enrich documents with context no chunk contains alone--navigation hierarchies with claim-evidence leaves, entity graphs, timelines, mind maps, ontologies, wikis, and memory. Compilation is not a better retriever: under a competent harness, structural artifacts gave no detectable accuracy gain on FRAMES, and a PageIndex-style outline degraded at scale. It is a supply layer. A navigation tree whose leaves are claim-evidence pairs--the mechanism we measure--cuts query tokens by 91% when evidence suffices and adds 1.9% when it does not, a 38% average reduction; this implies a 2% break-even sufficiency rate and a measured rate near 43%. A decision gate cuts latency from 140 s to 30 s, though substitution's token saving is bought with retrieval rounds and carries its own latency cost. Other artifacts--ontology, graph, wiki, timeline, mind map, memory--pay off in deployment (governance, integration, onboarding, temporal orientation, continuity) where accuracy benchmarks cannot see them. We describe RAGFlow's pipeline and harness (open source at https://github.com/infiniflow/ragflow), which turns the loop into five disciplined stations and keeps the context entering the model the smallest sufficient set; together they form an agent-native database: a store that hands the agent compiled, governed context instead of a ranked list of passages.

arXiv AI Papershace 14 días

TalentRL: Strategy-Guided Reinforcement Learning for Hierarchical Skill Routing

Natural-language skills enable large language model (LLM) agents to reuse procedures across tasks, yet most systems retrieve individual skills from a flat SkillBank without explicitly modeling the high-level strategy that governs their selection and use. We introduce TalentRL, a strategy-guided reinforcement learning framework for hierarchical routing over procedural memory. Coarse-grained talents represent recurring strategies for skill use, while fine-grained skills provide executable procedures. For each task, the policy model selects a talent that guides execution and softly prioritizes its associated skills during similarity-based retrieval. However, since task outcomes depend jointly on talent selection, retrieved skill context, and downstream execution, terminal rewards alone cannot isolate the contribution of the routing decision. TalentRL therefore trains talent routing with reward contrasts between rollouts of the same task under matched interventions on the talent-conditioned context, and optimizes downstream execution with group-relative advantages. Across seven open-domain and multi-hop question-answering benchmarks, talent-aware supervised fine-tuning improves average success rate by 7.53%, 6.71%, and 9.21% over the corresponding zero-shot baselines for Qwen2.5-7B, Qwen3-8B, and Llama-3.1-8B, respectively. Reinforcement learning further improves these SFT checkpoints by 7.58% to 9.32%, and TalentRL achieves the highest average success rate among all compared methods on all three backbones. On the agentic benchmarks ALFWorld and WebShop, TalentRL achieves the highest ALFWorld success rate (99.3%) and the highest WebShop score (90.7) among all compared methods.

arXiv AI Papershace 14 días

Confident Fools or Careful Searchers? A Behavioural Study of When LLMs Search Beyond Their Static Internal Memory

Agentic LLM assistants must repeatedly decide whether to answer from memory or to call a tool such as web search. Unnecessary searches cost latency and money; missing ones yield confident but unsupported answers. Prior work mostly measures verbalised confidence or trains new adaptive-retrieval methods, and rarely audits the search decisions that deployed products actually make. We introduce a benchmark of 400 multiple-choice questions that combines widely known inside-knowledge (IK) questions, which should be answered directly, with outside-knowledge (OK) questions about facts from news published after the models' training cut-offs (to our knowledge), which should trigger search or, without tools, abstention. Each item offers an abstain option, and two action-level scores, NAS and KIS, reward searching on OK questions and restraint on IK questions. Across five commercial LLMs with and without their vendors' web search, GPT-5 mini and Grok 4 Fast search on over 90% of OK but only 1% of IK questions, whereas Mistral Small searches on just 20% of OK questions, often abstaining instead of using its tool; one product's search indicator proved unusable. Searching is partly triggered by explicit temporal cues: on uncued OK questions, the NAS of the two selective models falls from about 0.72 to 0.57--0.58. The scores thus measure the search triggering of specific products, not metacognition in a strict sense.

arXiv AI Papershace 14 días

Differential rank preservation manifold learning

Manifold learning methods like t-SNE and UMAP preserve local neighborhoods, but they do not preserve the ranking of the neighbors inside them. Two embeddings can return the same neighbor set while disagreeing on which point is closest. This loss of order degrades downstream applications that value the order in the embedding. We propose Smooth Rank t-SNE (SR-tSNE), which adds a differentiable rank term to the standard t-SNE loss. To enable efficient training, this term uses a sigmoid rank approximation to explicitly force a point's true high-dimensional neighbors to outrank unrelated intruders and retain their original order. We also introduce Ordinal DLP (O-DLP), an extension of the Degree of Locality Preservation (DLP) metric that credits a neighbor only when it returns at its original rank. Across eight benchmarks spanning image, text, speech, remote-sensing and single-cell data, SR-tSNE outperforms 12 baselines in rank preservation and neighborhood overlap. Ablation studies demonstrate our rank term alone boosts O-DLP by 5% to 28%, a gain not achievable using standard pairwise loss. This advantage transfers to downstream applications. SR-tSNE achieves the best label-transfer performance across seven vision datasets. For cell-type discovery across six single-cell datasets, it ranks first in 17 of 18 dataset-metric combinations, whether clustering resolution is selected with or without cell-type labels. In intraday trading, SR-tSNE achieves higher returns, lower risk, and lower turnover than the baselines. To our knowledge, SR-tSNE is the first t-SNE-family method with explicit order preservation.

AWS Machine Learning Bloghace 14 días

Building intent-driven UI testing with Amazon Nova Act

Epsilon's Agentic Platform team delivers UI testing as a managed, multi-tenant service built on Amazon Nova Act and Amazon Bedrock AgentCore Browser Tool. Tests are written in plain English and run on browsers managed by AWS, giving every product team resilient, intent-driven testing with zero per-team infrastructure overhead.

arXiv AI Papershace 14 días

Elucidating the Design Space of LM Data Attribution

Data attribution methods for Language Models (LMs) are typically evaluated using downstream tasks like fact retrieval. However, existing evaluations vary multiple dataset properties simultaneously, making it impossible to isolate exactly why an attribution algorithm fails. To diagnose these failures, we explore the design space of LLM data attribution, with a primary focus on evaluation. We introduce the Controlled Attribution Benchmark (CATT) for fact tracing, in which we fine-tune LLMs on entirely fictitious knowledge, providing unambiguous ground truth for attribution. This synthetic setup allows us to systematically alter one dataset configuration at a time, such as document redundancy and lexical distractors. With this configurable evaluation, we demonstrate that redundancy and distractors consistently degrade attribution performance. Rather than proposing a new estimator, we examine two design choices that are underexplored: the loss used to compute gradients and how per-token attribution scores are aggregated into a document score. We find that the true causal evidence within a training document is highly localized, but standard methods dilute this evidence into noise by averaging attribution scores across the entire document. We find that replacing standard cross-entropy loss gradients with the negative log-odds gradients surfaces the critical tokens from other uninformative ones. We can then aggregate only the top token scores to improve ground truth recall by 35 to 72 percentage points across multiple baselines on our diagnostic benchmark. The negative log-odds correction also generalizes to FTRACE-TREX, improving counterfactual retraining performance for every method we test, by up to 4.9x.

arXiv AI Papershace 14 días

Probing Spatial State Representations in Visual Foundation Models

An observer sees only part of the surrounding world, yet must keep track of what lies beyond the current view. Spatial perception therefore involves more than recognizing individual images: it calls for a persistent spatial state that is updated under self-motion and used to guide action. Do visual foundation models provide representations that support such a state? We study this question through four levels of spatial capability---spatial structure, object persistence, predictive spatial update, and actionability---using frozen-feature probes on ten representations under streaming egocentric observations of static indoor environments. The resulting capability profiles vary across tasks, network layers, and prefix lengths: strong readout of the visible scene does not ensure accurate object persistence or future-feature retrieval. Matched controls separate readout accuracy from reliance on visual history and motion. Native future prediction retains recoverable spatial structure, but a spatial reader trained on real features does not transfer to predicted features unchanged. Finally, goal-dependent candidate selection does not establish reliable continuous action utility: predicted actions respond to goal changes, yet supplying the correct goal does not reliably improve progress. Across the four levels, object persistence and actionability build on camera-pose readout, whereas feature prediction builds on visible-scene readout. Together, these findings motivate evaluating spatial representations not only by what can be decoded from them, but by whether that information is maintained beyond observation, updated predictively, and used for goal-directed action.

arXiv AI Papershace 14 días

Testimony from Timelines: Evidence-Grounded Significance Testing of Temporal Clinical Hypotheses

Ask a language model whether sales tend to dip after a product recall, or whether one side effect tends to follow another after a vaccine, and it answers in a fluent sentence that reads like a finding, with no test behind it. Assertions about how events unfold across many cases fill the Web, and each is a temporal hypothesis that only a significance test over the right records can settle. The verifier must specify the null, select the test and decide which records bear on the hypothesis, and any of these choices can go wrong while the conclusion still comes out right. Yet benchmarks for language-model data analysis grade only the conclusion, with the relevant data handed over. We study the problem in health, where a wrong verdict is costly, and present STRATA, a benchmark that grades every decision on the way to a verdict: 8,735 hypotheses over the clinical timelines of vaccine adverse-event reports, with gold plans, evidence sets, p-values and verdicts derived under a released codebook and reproduced by a deterministic oracle. We also propose ART (articulate, retrieve, test), a staged paradigm that exposes each decision as an inspectable output. Extensive experiments with five model configurations show that systems fail mostly at choosing which records count, a verdict-evidence gap that accuracy alone conceals. ART closes most of this gap and brings a 31B open model to within 0.01 of the frontier models under the same paradigm; the frontier models succeed once given the codebook.

arXiv AI Papershace 14 días

When to Assist and What to Suggest: Joint Gating and Memory-Grounded Hint Generation in One Compact Decoder

Proactive assistants offer short, memory-grounded hints during a conversation, deciding when to assist and what to suggest. Current systems split this between a gate and a larger generator; we evaluate both together on 28,058 decision points. Stage-wise generation scores are oracle-gate ceilings: behind a real gate, end-to-end exact match is 0.2619 against a ceiling of 0.2971, and end-to-end precision is 0.2387, leaving substantial headroom. A gate pays off through precision and compute, whose saving falls from 4.39x with an 8B generator to 1.38x with a 1B one. The memory profile drives hint quality: removing it costs 24.7 exact-match points (rescaled BERTScore 0.57 to 0.35), while retrieval and a larger generator add nothing on exact match. One 1B decoder can do both, gating from an intermediate layer and generating in the same pass. At its final checkpoint it matches or exceeds the two-model pipelines on automatic end-to-end measures with 1.31-1.76x less wall-clock time than the matched 1B pipeline; across three seeds, its deeper exit ties the separate gate. A human evaluation (1,144 ratings over 32 dialogues) supports the memory effect and rates the single model level with the matched 1B pipeline and below the 8B one. Graded metrics and LlamaPIE's native judge track human ratings more closely than exact match, which we report as a strict lower bound.

arXiv AI Papershace 14 días

Advancing Clinical Genomic Analysis with a Skill-Centred Agent

Clinical genomic analysis requires experts to integrate evolving guidelines, databases, and computational evidence. As sequencing has become increasingly accessible, this complex and labor-intensive process has become a major bottleneck to the timely use of genomic information in clinical decision-making. To fill this gap, we developed ClinGeX, a large-language-model agent that composes 105 curated executable skills through task-specific retrieval and dependency-aware orchestration. On 223 temporally held-out benchmark cases, ClinGeX achieved 71.7\% macro-averaged task-level concordance with expert reference classifications, compared with 45.7\% for the strongest comparator. In a three-condition, counterbalanced human-AI collaboration study, 45 clinical genomic professionals assessed 135 clinically derived cases under routine (no AI assistance), DeepSeek-assisted and ClinGeX-assisted conditions. ClinGeX assistance increased raw case-level concordance across criterion-level variant analysis, genomic interpretation in rare disease and molecular oncology decision-making (51.1\%, 57.8\%, 64.4\% versus 15.6\%, 31.1\%, 20.0\% in comparator conditions) and reduced reported interpretation time in all scenarios. During an exploratory one-month real-world non-interventional assessment, 11 experienced professionals submitted feedback on 96 self-selected analysis episodes and reported frequent conclusion adoption and time savings. These findings support executable domain skills as a promising approach to augment specialist genomic interpretation.

arXiv AI Papershace 14 días

Retrieval as Traversal: Value-Indexed Memory for Stateful Long-Context Agents

Long-context models repeatedly solve closely related retrieval problems. Attention or global retrieval can retain large stores, but each forward pass again searches for what matters. Recurrent compression avoids this search only by deciding earlier what information to preserve. Neither mechanism directly exploits that useful memories at neighbouring retrieval steps may substantially overlap. We propose an alternative in which retrieval is a persistent local traversal through an address space. We separate a memory's content from its address. Utility is defined by the information revealed by retrieving the content, while geometry determines only how cheaply that information can be reached. This leads to a cost-regularised indexing objective that preserves information value inside local neighbourhoods while charging for every memory made locally accessible. Its marginal-value form gives both attractive and exclusion pressures. Thus, complementary memories should be co-retrievable when the additional value exceeds retrieval cost, whereas redundant memories should not occupy the same retrieval set solely because they are individually useful. As a proof of principle, we test only the simplest fixed-geometry instance on MiniGrid Memory. A persistent local traversal head reaches 0.95 mean return after 0.59M training steps with a five-token context, compared with 0.94M for Gaussian absolute addressing and more than 2M for categorical absolute addressing. The experiment does not test learned geometry or repeated retrieval. Instead, it motivates the hypothesis that stateful retrieval may be easier to learn than stateless. To test the full proposed method further experimentation is necessary.