Sección

Investigación

Papers y preprints de los feeds de investigación que seguimos.

6415documentos

FLEET: From Logits Entropy to Enhanced Trajectories in Text Generation

Solutions based on large language models (LLMs) often rely on temperature sampling to improve accuracy and stability by aggregating multiple samples from the completion distribution. However, this memoryless approach is inherently suboptimal: because it lacks awareness of prior generations and their evaluations, it produces an increasing proportion of semantically duplicate answers as more samples are drawn, leading to diminishing returns. To address this limitation, we introduce FLEET, a novel method that integrates a memory mechanism into the generation process. FLEET represents each generation as a sparse trajectory through states whose entropy exceeds a predefined threshold and uses these trajectories to infer per-token utility scores that adjust the logits. Benchmark evaluations demonstrate that FLEET achieves the same accuracy as the repeated sampling baseline, with a 3x speedup, and substantially improves accuracy on complex coding tasks (LiveCodeBench Pass@32 increases from 59.9% to 66.2%) under the same budget. Furthermore, in the greedy-decoding configuration evaluated here, the approach is deterministic and uses a single calibration pass to derive its principal hyperparameters, requiring only minimal modifications to existing LLM pipelines.

arXiv AI Papershace 2 meses

InternW0: A Foundational Physical World Model for Efficient Real-World Interactions

Physical intelligence requires more than predicting how the world may evolve: predictions must remain actionable as the world continues to change. We introduce InternW0, the first instantiation of the InternW physical world model series from Shanghai AI Laboratory, built around omnimodal interfaces, asynchronous multi-frequency processing, and local physical modeling under partial observations and external influences. InternW0 jointly learns future visual dynamics and continuous robot control through an asymmetric video--action architecture with flow matching. A high-capacity video expert provides longer-horizon predictive context, while a lightweight action expert operates at a faster timescale. Instead of regenerating the future for every action update, InternW0 reuses layerwise K/V and adapts it to newly observed states through observation-conditioned context routing. Domain-specific interfaces and soft prompts support heterogeneous embodiments, while contact-aware post-training incorporates force and tactile signals for contact-rich manipulation. We train InternW0 on approximately 7,200 hours of heterogeneous robot and egocentric data, including EgoLab, a 275-hour real-laboratory egocentric dataset. Evaluation spans simulation benchmarks and real-world scientific tasks, including a 15-stage metal--organic framework synthesis workflow and 5-stage contact- and force-aware dexterous manipulation for general-purpose quantitative pipetting. These results advance scalable, asynchronous, and science-native physical world models for universal and efficient real-world interactions.

arXiv AI Papershace 2 meses

FedIncome: Federated Learning for Income Estimation in Digital Lending Under Data Sovereignty Constraints

Verified income is often unavailable in digital loan applications, forcing lenders to rely on reported income and potentially leading to over-lending, overly conservative offers, or rejection of creditworthy applicants. Cross-institutional data-sharing constraints make this problem especially difficult for smaller lenders with limited training data. We introduce FedIncome, a federated learning framework for income estimation that enables institutions to train a shared model without pooling raw borrower records. Using more than one million LendingClub loans partitioned into 50 state-level clients, we simulate a heterogeneous lending consortium. The best federated model achieves out-of-time R^2=0.608, compared with 0.619 for a pooled centralised benchmark. Small-sample clients obtain an average out-of-time R^2 improvement of 3.8 percentage points relative to the pooled centralised benchmark, while the fitted client-level relationship places the empirical crossover at approximately 4,790 training observations in this setting. When pooling is infeasible and the relevant alternative is local-only training, federation improves out-of-time performance across all sample-size groups, with the largest gains for data-scarce clients. We also combine federated income estimates with state- and income-specific debt-to-income thresholds. In a retrospective decision analysis, replacing reported income with the federated estimate increases simulated approval rates with only modest changes in observed default rates. FedIncome supports collaborative learning under data-locality constraints with little aggregate loss relative to pooled training and larger gains relative to local-only estimation.

arXiv AI Papershace 2 meses

Brain-to-Language Decoding: Tasks, Signals, Methods, Evaluation, Practical Use and Beyond

Brain-to-language decoding translates neural activity associated with language production, internal speech and perception into linguistic or expressive outputs. It offers a route to restoring communication after speech loss and a means of studying how the brain represents language. Advances in neural recording and representation learning have expanded the field from constrained recognition and acoustic reconstruction to text generation, streaming personalised speech and facial animation. This survey synthesises these developments across invasive and non-invasive measurements, drawing on a search without a lower year limit and source-led updates through September 2026. We connect Articulated, Inner and Perceived tasks to the neural populations they engage, the representations available to decoders and the outputs those representations can support. We examine model development, public resources and the evolution of evaluation, and compare published performance and communication costs within their reported protocols. The synthesis identifies complementary routes to progress: phonetic, acoustic and semantic targets preserve different aspects of a message; shared representations support reuse across recording conditions and tasks; and online communication increasingly depends on calibration, feedback and user control alongside decoding accuracy. Shared benchmarks enable algorithmic comparisons, while longitudinal studies reveal the demands of sustained use. We discuss these developments and their remaining limitations, then outline a prospective five-level trajectory from commands and language to meaning, scenarios and bidirectional cognitive exchange

arXiv AI Papershace 2 meses

Learning Local Heterogeneity and Cross-Region Context for Large-Scale Traffic Forecasting

Traffic flow forecasting is essential to intelligent transportation systems. Large-scale traffic forecasting requires jointly modeling local spatial dependencies and cross-region context.Spatial dependencies between geographically neighboring nodes are heterogeneous due to differences in road identity and travel direction, while acquiring global information through allpairs node interactions incurs substantial computational costs. Therefore, capturing local heterogeneity while efficiently acquiring long-range context remains an important challenge in largescale traffic forecasting. To address these challenges, we propose LoReST, a Local-Region Spatial Temporal network that models spatial dependencies at two complementary granularities: node neighborhoods and road network regions. Specifically, relation-aware local aggregation captures heterogeneous dependencies within geographic neighborhoods through road and direction specific feature transformations. Cross-region interaction constructs region representations through mean pooling, exchanges long range context via inter-region attention, and broadcasts it back to nodes. By integrating local information aggregation with crossregion interaction, LoReST is able to effectively achieve spatial dependency learning in large-scale road networks. Experiments on four datasets of the LargeST benchmark show average relative reductions of 4.78%, 3.60%, and 5.75% in MAE, RMSE, and MAPE, respectively.

arXiv AI Papershace 2 meses

Pheno-GS: Phenoscape-scale Geodesic Sinkhorn

High-throughput single-cell data is now collected across large patient cohorts. Understanding patient-level heterogeneity from cellular-level data motivates phenoscaping: embedding each single-cell distribution as a "datapoint," with distances given by optimal transport (OT). Computing geometry-aware OT at this scale, between all pairs of patient datasets, remains an open challenge, since existing methods either rely on Euclidean ground metrics that distort manifold structure or fail under sparse, unevenly sampled, or large-scale data. We present Pheno-GS (Phenoscape-scale Geodesic Sinkhorn), which computes accurate, scalable geodesic transport distances under noisy, unbalanced, large-scale settings via three components: (1) graph connectivity regularization for well-defined geodesics on sparse/disconnected manifolds; (2) an unbalanced OT formulation via KL marginal penalties; and (3) a batched matrix algorithm computing all pairwise distances in one heat diffusion (over 200 faster than Geodesic Sinkhorn for 500 distributions). We validate Pheno-GS on synthetic benchmarks and a CyTOF perturbation dataset.

arXiv AI Papershace 2 meses

Compliant AI Infrastructure for Regulated Finance: A tiered multi-agent framework with DLT audit trails for financial operations in DACH

We present a compliance-first architecture for AI in regulated finance that treats regulation as an orientation layer rather than a deterministic ruleset. A matrix of regulatory intent and exposure provides a compact classification handle, which a governed policy compiler then maps into concrete prohibitions, obligations and runtime budgets. Prohibitions constrain feasibility and block externalisation, while obligations extend tasks with artefacts that must meet explicit admissibility criteria. Committee activation remains policy-driven and proportionate, preserving efficiency while ensuring supervisory oversight. Evidence, decisions and reason codes are bound to a permissioned DAG with deterministic timestamping, enabling replay, provenance checks and clear attribution of failure. Clause-level legal indexing with effective dates and capability-based agent routing ensure portability across DACH and the wider EU. The result is assurance by construction: compliance is embedded in execution and verifiable by auditors without sacrificing proportionality or transparency.

arXiv AI Papershace 2 meses

SHRAV: State-Hypothesis-Reason-Action-Verify Framework for Physical Modeling and Inverse Design

Physical modeling and inverse design require computation that can continue from reusable state. We introduce SHRAV, an architecture-independent computational framework organized around State, Hypothesis, Reason, Action, and Verify. Its central mechanism is a state-continuation core with declared reuse boundaries and explicit roles for learned evolution and numerical quantities. Forward configurations evolve predictive state and read out physical responses; inverse-design configurations additionally generate target-directed modifications and consume evaluator feedback. Electromagnetic world-model studies are mapped to forward configurations, with selected readout and reuse diagnostics reported here. Computational lithography demonstrates an inverse-design configuration: four fixed-weight design updates improve thresholded aerial-image intersection-over-union from 0.5313 to 0.8153 under independent scalar-pupil replay, with maximum absolute prediction-replay difference approximately 0.000824 between predictor estimates and independent replay.

arXiv AI Papershace 2 meses

InGuard: Towards Generalized Inner Guardrail for Safe Text-to-Image Generation

Modern text-to-image (T2I) models generate high-quality images from arbitrary user prompts, yet they can just as easily produce not-safe-for-work (NSFW) content. Conventional outer guardrails consist of two components: a prompt classifier that checks for risk before generation, and a post-hoc image classifier that checks the fully generated image. In this design, both classifiers operate outside the generation pipeline and do not use the model's own representations. This separation can limit prompt-screening accuracy, while the image-side check runs only after the full generation cost has been spent. Moreover, a flagged prompt can only be rejected, even when it could be adjusted to produce a safe image. In this work, we propose the Inner Guardrail (InGuard), a safety framework that works inside the pipeline on the model's own representations, leaving base-model parameters untouched. First, a risk classifier grades each prompt as unsafe, risky, or benign based on the text encoder's embeddings, with no external language model. Second, SAGE (Soft-gated Asymmetric Guardrail for Embeddings) modifies the embeddings of risky prompts, aiming to return a safe image instead of a refusal. Third, a latent detector checks the one-step clean latent estimate midway through denoising, reaching nearly image-level performance and halting generation when risk is detected. We also construct the RevGen Safety Benchmark to evaluate T2I safety under realistic conditions: 10,000 prompts built through real-image reverse generation, with a rewriting step that supplies controlled intellectual-property (IP) characters, covering graded porn/gore risks, categorical IP risks, and benign negatives. Across five open-weight T2I models, InGuard reaches 97.9-98.8% safety rate, matching or exceeding the outer guardrail, with 57.5-73.5% less benign disturbance, ~3.7x fewer parameters, and 50-55.6% of denoising steps skipped.

arXiv AI Papershace 2 meses

BiCFlow-MER: Orchestrating Discriminative and Generative Multimodal Emotion Recognition via Conditional Transport

In multimodal emotion recognition (MER), human affective states are inferred by integrating complementary cues from multiple modalities. In audio-text MER, affective cues are often entangled with speaker style and lexical content, while cross-modal disagreement further complicates how the evidence should be integrated. Under conventional discriminative fusion, multimodal evidence is compressed into a terminal prediction, with modality-specific cues and conflict information insufficiently preserved. In large generative affective models, by contrast, affective reasoning is typically embedded in language decoding, leaving emotion evidence implicit and difficult to verify in a structured space. To address these limitations, BiCFlow-MER (Bidirectional Conditional Flow for Multimodal Emotion Recognition) is proposed as a conditional-flow framework in which audio-text MER is formulated as generative evidence transport within a structured emotion space. Within BiCFlow-MER, emotion-oriented evidence is disentangled from speaker-style and lexical-content factors to construct a conflict-aware affective condition. Guided by this condition, each utterance is transported to an explicit emotion-space endpoint through a bidirectional rectified flow. Candidate emotions are jointly verified through adaptive prototype-cloud scoring of the transported endpoint and backward class-to-condition consistency with the original multimodal condition, enabling conflict-aware recognition. BiCFlow-MER is shown to outperform all compared methods across IEMOCAP, MELD, and the zero-shot CASE benchmark. By orchestrating discriminative recognition and generative evidence modeling through conditional transport, BiCFlow-MER defines a new MER paradigm.

arXiv AI Papershace 2 meses

Can Jev Judge Radiology Reports? Evaluating a System One Model for Clinical Factuality

An AI-generated radiology report can resemble a physician's report while omitting an abnormality, adding an unsupported finding, or reversing its presence. Measuring these factual differences is essential for evaluating report generators. We study Jev, a System One decision model, as a simple, low-cost judge of agreement with physician-written reference reports. Our evaluator checks whether each statement is supported by the other report and combines these judgments in both directions to capture unsupported claims and omissions. A single-question configuration reaches Kendall correlations of 0.573 on RadEvalX and 0.398 on RadEvalExpert with expert error counts, outperforming an open natural language inference judge under matched decomposition and aggregation. One support question per statement retains similar expert agreement to seven while using 43-45% fewer judgment input tokens. At the documented API price, judgments cost under three cents per hundred report pairs, excluding local decomposition. In a separate controlled-error test, Jev detects false negation with an AUROC of 0.977. Local RadMatch achieves stronger agreement on clinically significant errors in both expert datasets and on total errors in the shared RadEvalExpert subset. Finding-count and error-scope analyses show that benchmark agreement reflects report size and error definitions as well as medical error detection. These results support Jev as a practical judgment component for measuring factual differences in generated radiology reports and identify where more elaborate evaluation remains valuable.

arXiv AI Papershace 2 meses

State-Grounded Conditioning: Wrapping User-Facing LLM Agents Where Direction Depends on Live State

We introduce State-Grounded Conditioning (SGC), a design principle for user-facing LLM agents that must condition on live user state (game state, session history, live inventory), and a distinct failure class we call direction drift: task-complete responses whose chosen direction misaligns with the current state. SGC externalises state-dependent control into rule kernels over structured inputs and three primary state slices, via Perception, Grounding, and Interaction wrappers with explicit conditioning dependencies. We evaluate SGC on a 200-session anonymised benchmark (1,000 assistant model turns) from an in-game conversational coaching agent that guides players through consecutive competitive matches, reporting mean first-token latency and five human-annotated dialogue-quality metrics that jointly cover factual grounding and coach-like guidance progression. The Perception wrapper holds mean first-token latency at 1.5s (vs. 6.1s for PE-Agent inside a production tool-use harness); enabling all three wrappers lifts turn-level grounded accuracy from 61.1%/69.8% (Prompting / PE-Agent) to 96.7% and session-level grounded accuracy from 20.0%/26.5% to 83.5%; session-level grounding-failure incidents drop by 78% relative to the strongest baseline. A cumulative ablation shows complementary incremental gains as the wrappers are added. These results inform approximate state-slice orthogonality, without establishing independent per-wrapper effects.

arXiv AI Papershace 2 meses

When Context Misleads: In-context Learning with Jurisdiction in Large Language Models

In-Context Learning (ICL) has become a cornerstone of modern LLM deployment. However, existing ICL post-training methods have a critical blind spot: they excel at extracting patterns from demonstrations while often neglecting context authority, the ability to determine whether contextual information should govern the final answer. To benchmark this capability, we introduce FakeContextBench, which contains pseudoscientific claims across seven domains. Our evaluation of commercial and open-source models shows that large-scale pre-training alone is insufficient for reliable context-authority discrimination. Moreover, prevalent ICL fine-tuning methods can increase susceptibility to misleading context, reducing reality accuracy by up to 14.95 percentage points relative to the base model. To address this trade-off, we propose Jurisdiction In-Context Learning (J-ICL), a post-training framework that incorporates context validation into the training objective. Across four model backbones, J-ICL improves ICLEval by an average of 5.84 percentage points and reality accuracy by 9.20 points over the corresponding base models. It also raises the Reality Rate by an average of 18.09 points relative to MetaICL and Symbol Tuning. These results demonstrate that ICL capability and resistance to deceptive context can be improved together. The benchmark is available at https://github.com/peilin717/FakeContext-Bench.

arXiv AI Papershace 2 meses

Efficient Linear Bandits via Cluster-Aware Sketching

We study the problem of computational efficiency for linear bandits in high-dimensional settings with a finite arm set. In linear bandits, the increase in the dimension d of the feature vectors leads to growing computational costs of O(d^2) at each round of update. Traditional sketching-based methods such as SOFUL reduce computation via fixed-size matrix sketching, yet run the risk of incurring vacuous linear regret when the spectral tail of the data is heavy and the sketch size is inadequately selected. To guarantee regret convergence and effectively reduce computational costs, we introduce a clustering mechanism and propose the Cluster Sketch Linear Bandit (CS-LB) algorithm. Our method preserves the full covariance information in each cluster to guarantee robust sublinear regret without spectral-tail vulnerabilities, performs cluster switching by assigning a sentinel for each cluster, and reduces per-round update computation to O(l^2d) via a tunable sketch size $l

arXiv AI Papershace 2 meses

Hidden not Deleted: How Networks Suppress Entangled Features

Concept erasure methods that operate via linear projection assume that features occupy separable subspaces. We show this assumption fails under dense superposition: when two features are forced into an antipodal pair sharing a single subspace, state-of-the-art linear erasure destroys both, not just the target. Networks trained with gradient descent instead solve this problem non-linearly, but not uniformly: they converge to one of two distinct circuit-level solutions depending on initialization, which we call mirror and shadow solutions. We map this bifurcation as a function of feature entanglement, show it reflects a stable attractor structure rather than an artifact of our setup, and use targeted causal interventions to demonstrate that both solutions leave a substantial, measurable trace of the erased feature's representation intact, recoverable through a single scalar patch rather than requiring any further training. This mirrors a failure mode recently observed empirically in LLM unlearning, where suppression rather than deletion allows forgotten knowledge to resurface; our results offer a mechanistic, causally-validated account of why that failure mode occurs.

arXiv AI Papershace 2 meses

MWE-ECL: Recoverable Long-Range Context Does Not Always Override Local Lexical Priors

Long-context evaluations often test whether a model can recover distant evidence, but recoverability does not guarantee behavioral influence. We test the prediction that a distant discourse anchor can remain explicitly recoverable yet fail to change the locally preferred reading of a familiar multiword expression; such failures should concentrate when the model's no-anchor default conflicts with the anchor, while prior-correct decisions remain largely preserved. We introduce Multiword Expression Effective Context Length (MWE-ECL), a bilingual diagnostic whose matched anchor-retrieval, no-anchor prior, and interpretation prompts measure explicit recoverability, model-observed defaults, and anchor-conditioned decisions, respectively. Across eight English deployment panels on a shared 0-128K grid, retrieval-control accuracy on prior-conflict items is 0.989-1.000, prior-conflict override spans 0.806-1.000 (0.809-1.000 after conditioning on correct retrieval), and preservation of prior-correct decisions remains 0.977-1.000. A same-call control querying retrieval and interpretation in one prompt reproduces the gap for DeepSeek V4 Pro (1.000 retrieval versus 0.900-0.920 interpretation), showing that separate invocations are not its sole explanation; smaller or absent gaps in the other two models bound its generality. For DeepSeek V4 Flash, separate prompt-fit tests retain perfect retrieval with lower interpretation at 512K and 1M, while foil-consistent cues shift the no-anchor prior far more than retrieval; cross-model cue effects are heterogeneous. A separately reported 10-family Chinese subset shows similar descriptive gaps, but imperfect retrieval for some models prevents an integration-only attribution. MWE-ECL therefore evaluates whether explicitly recoverable distant context changes a competing local semantic decision.

arXiv AI Papershace 2 meses

The Capability Manifold and ML Scaling Laws

Existing machine learning (ML) scaling laws relate predictive loss to compute, model parameters, and data. However, as models are increasingly deployed through agentic harnesses, loss alone is insufficient to characterize downstream performance: models with similar loss can exhibit different capabilities in reasoning, retrieval, planning, and adaptation. Yet, no unified framework connects such capabilities to the coupled resources available across the ML lifecycle. We bridge this gap by introducing a capability manifold, a multidimensional framework mapping downstream capabilities to pre-training, post-training, and test-time resources through bounded scaling functions. Analytical Jacobians quantify capability sensitivity to resource changes and interactions. As an initial application, we embed Kaplan- and Chinchilla-type scaling laws and test-time compute within the framework, demonstrating how existing scaling relationships can be unified as trajectories on a common capability manifold.

arXiv AI Papershace 2 meses

Does Step Law Transfer to Small-Scale Language Models? An Empirical Recalibration Below 59M Parameters

Step Law gives power-law formulas for the optimal peak learning rate eta* and batch size B* when pre-training language models. It was calibrated on models between 59M and 1B parameters; the small-model regime N < 59M was never tested empirically by its authors. This regime matters for single-GPU training, interpretability research, educational experiments, and settings where larger models are infeasible on memory or cost grounds. We test whether Step Law transfers to small language models. We consider three outcomes: H1, the original coefficients work directly; H2, the power-law form holds but with different coefficients; and H3, a power law does not describe the optima in this regime. All experiments use a single nanoGPT/TinyStories pipeline with a 2048-token BPE vocabulary, AdamW, and a warmup-cosine schedule. The optimum for each (N, D) cell is extracted from the loss surface L(eta, B) via a local quadratic approximation in log-log coordinates over the smoothed training loss. The final dataset contains 29 unique (N, D) cells and 935 analysis-ready runs. The main refit uses 25 cells (815 runs) in the working range 4 <= D/N <= 600. On the pooled data we accept H2: the functional form is preserved, but the coefficients differ from the original. We obtain eta*(N, D) = 0.0985 N^(-0.508) D^(0.238) (R^2 = 0.834) and B*(D) = 3.6 x 10^(-4) D^(0.931) (R^2 = 0.950). Step Law's structural claim that B* is independent of N is reproduced (p = 0.87), but the growth of B* with D is nearly twice as steep as in the original work. Direct transfer of Step Law systematically overestimates the optimal learning rate: the median ratio eta_SL / eta* is approximately 4.0x, with a range of 2.4x to 6.6x.

arXiv AI Papershace 2 meses

VCMM: Variance-Calibrated Momentum for Multimodal Learning

Multimodal joint training often suffers from modality imbalance, where a dominant modality suppresses the optimization of others. Existing methods mainly balance modality learning by modulating gradient magnitudes or directions, modifying optimization objectives, or adjusting training strategies, with most interventions focusing on the current update. However, when combined with widely used momentum-based optimizers, the update also incorporates accumulated information from previous gradients, which is not explicitly addressed by current-step modulation alone. To address this issue, we propose Variance-Calibrated MomentuM (VCMM), which adapts gradient memory to modality-specific gradient dynamics. Specifically, VCMM estimates minibatch noise and temporal drift online and uses their relative strength to determine modality-specific momentum through a Kalman-inspired controller. We further center the control signal across modalities and apply exact bias correction for the time-varying first moment, enabling adaptive gradient memory without extra network passes or explicit learning-rate scaling. Experiments on four multimodal benchmarks demonstrate consistent improvements with modest training overhead.

arXiv AI Papershace 2 meses

DCRL: Decoupling and Coupling Reinforcement Learning via Policy-Reward Manifold Alignment

Reinforcement learning (RL) has emerged as a key paradigm for improving the reasoning capabilities of large language models (LLMs). However, existing reward systems, such as rule-based and reward-model-based, often exhibit issues such as unstable optimization and reward hacking. In this work, we revisit the general reasoning of LLMs from a geometric perspective, conceptualizing it as a coupled manifold composed of three interdependent sub-manifolds: logical deduction, evaluation, and representation. Based on this perspective, response generation in RL can be interpreted as a decoupling process from the evaluation manifold, while reward estimation corresponds to a decoupling process from the logical deduction manifold. The limitations of rule-based and reward-model RL systems can be geometrically interpreted as the mismatch of policy-reward manifolds during RL process. To address the aforementioned misalignment, we propose Decoupling and Coupling Reinforcement Learning (DCRL) framework, which incorporates two key components: (1) a syllogistic logic-based prompt evolution mechanism that dynamically refines reward rubrics to enhance the expressiveness of the reward manifold; and (2) a policy-reward re-coupling mechanism that jointly updates the reward and policy models, ensuring consistent evaluation and mitigating manifold mismatch during training. Theoretical analysis and extensive experiments across multiple reasoning domains demonstrate that DCRL consistently outperforms both rule-based and reward-model baselines. Notably, a Qwen3-4B model trained under DCRL surpasses a Qwen3-32B baseline and approaches the performance of a Qwen3-235B model, highlighting superior effectiveness and generalization in RL.