Tema

Código abierto

1634documentos

158últimos 7 días

When Credit Turns Against Tools: Transition-Aware Credit Assignment in Agentic Reinforcement Learning

Agentic reinforcement learning with verifiable rewards trains language models through multi-turn tool interactions, yet their reasoning performance and tool-calling ability can decline during training. We trace this degradation to three coupled mechanisms: (1) tool-token imbalance produces negative overall tool credit when incorrect trajectories contain more tool tokens per turn than correct ones; (2) aggregation amplification can further strengthen this negative signal as token-mean aggregation gives longer incorrect trajectories greater weight in policy updates; and (3) failure-driven retries prolong trajectories as models tend to call tools again after execution errors, extending potential negative credit to recovery attempts. To counter this suppression, we propose Transition-Aware Credit Assignment (TACA), which targets the transition from reasoning to tool use through supervision of both call initiation and generation. Specifically, TACA uses Advantage Projection to enforce a nonnegative mean advantage on tool-call tokens within each prompt group through a minimal uniform shift. Additionally, TACA guides tool-use decisions with Decision Contrast, comparing rewards for tool-assisted and direct answers to the same prompt. Through Supervised Replay, it reinforces tool-call generation using calls from previously sampled correct trajectories. Across models and tasks, TACA improves performance and sustains more reliable tool interaction during training, raising mean accuracy over MT-GRPO on six reasoning benchmarks by 4.61 percentage points for Qwen2.5-1.5B and 3.61 for Qwen2.5-3B. Our code is available at https://github.com/tmlr-group/TACA.

arXiv AI Papershace 4 días

PROLA: Principal-Orthogonal Low-rank Adaptation for Predictive Spatiotemporal Weather Downscaling

High-resolution weather prediction is important for resolving local atmospheric patterns missed by coarse global forecasts. Large pretrained weather backbones have improved global forecasting, but adapting them to produce future high-resolution fields from recent coarse states remains challenging and costly. We formulate this setting as Predictive Spatiotemporal Weather Downscaling (PSWD), a task that jointly refines spatial and temporal resolution while predicting future high-resolution trajectories. To this end, we propose a framework that decomposes prediction into spatial and temporal stages using a shared pretrained backbone, fixed numerical scaffolds, and lightweight residual heads. For backbone adaptation, we introduce PROLA (PRincipal-Orthogonal Low-rank Adaptation), which allocates a fixed low-rank budget between pretrained weights' principal directions and their orthogonal complement. PROLA further rescales rank-1 optimizer updates using gradient signals, improving adaptation without increasing trainable rank. Experiments in the PSWD setting show that PROLA outperforms representative low-rank adaptation baselines. Further experiments across diverse datasets and tasks support the broader applicability of PROLA, showing that its benefits extend beyond the primary weather downscaling setting.

Hacker News (AI)hace 5 días

Show HN: LLM weight editing tool made by a 15-year-old

en▲ 5

arXiv AI Papershace 5 días

Hybrid Source Code Change Impact Analysis

Change impact analysis (CIA) helps developers identify code entities impacted by their initial changes. While several approaches for CIA were proposed, they still struggle with either low precision or low recall, thereby increasing manual effort for developers to verify the results. Furthermore, limitations in the used datasets and experimental setups can bias the results reported by many CIA approaches. In this paper, we first address the dataset limitations by manually labeling 1,931 changed code entities stemming from 263 commits of 22 open-source Java projects. Next, we introduce two variants of katIA, our code change impact analysis approach that combines static impact analysis with the reasoning capabilities of large language models (LLMs). The first variant uses a single prompt (katIA-SP) for refining an initial static impact set while the second variant uses three specialized agents (katIA-MA). The evaluation with our curated dataset shows that katIA-SP with Gemini 2.5 Flash achieved an F1-score of 22.4% when matching impacted entities by their fully qualified names (FQN) and 27.0% when matching them by their class and entity names (CE). The latter shows an improvement of 2.0 percentage points over the state-of-the-art. This score is further improved by katIA-MA which, using Gemini 2.5 Flash, yields an FQN-matching F1-score of 29.1% and a CE-matching F1-score of 33.3%. While these improvements come at the cost of increased median generation time from 0.570 s to 11.214 s and median token usage from 1,235 to 6,462 tokens, the improved performance justifies the additional cost. The comparison of three LLMs using katIA-SP finds that Gemini 2.5 Flash is the most consistent model. These results indicate that combining static impact analysis with LLM-based refinement, particularly with the multi-agent variant, can improve code change impact analysis.

arXiv AI Papershace 5 días

Interpretable Cyber-Physical Attack Detection in Smart Grids: A Leakage-Controlled Multi-Testbed Evaluation of Fusion, Robustness, and Observability Limits

Cyber-physical smart-grid detectors are typically treated as black boxes and validated exclusively on clean, temporally shuffled datasets, which risk overly optimistic performance claims. This work presents a multi-testbed evaluation study built around an interpretable, time-windowed fusion Random Forest, adopting the pessimistic result whenever evaluation protocols disagree. (i) On a public paired cyber-physical dataset, shuffled cross-validation is optimistic: macro-F1 falls from 0.96 to 0.88 +/- 0.06 under rolling-origin temporal evaluation. A topology-free graph convolutional network (GCN) serving as a control confirms that this lightweight model is competitive on these splits. (ii) The adaptive and adaptive-joint feature-space attacks, which are infeasible as realistic grid manipulations, degrade macro-F1 to 0.67 and 0.34, respectively. Adversarial training and randomized smoothing partially restore performance, and a physics-constrained evolution-strategy evasion on the IEEE 14-Bus testbed shows the same vulnerability pattern. (iii) On real IEC-104 traffic, the cyber view reliably catches structural attacks (recall of 0.87 to 0.98) but largely fails on semantic value manipulation (recall of 0.11 to 0.35). An observability taxonomy on the IEEE 14-Bus system reveals that stealthy false data injection defeats all three observation views. (iv) On a full AC weighted least-squares state-estimation testbed, a forecasting-aided temporal test is evaded by slow ramps precisely where a rate-invariant physics-manifold test sustains detection at a common 5% false-alarm target. A full end-to-end evaluation on IEEE 118-Bus confirms the ordering but bounds its strength: the manifold test detects every trial down to 0.05 degree/step and 0.48 at 0.02 degree/step, against 0.14 for the temporal test. All code and benchmarks are released.

arXiv AI Papershace 5 días

How Short Is Too Short? Length-Conditional Calibration and Abstention for Detecting AI-Generated Text

Detectors of AI-generated text are usually tuned and tested on passages of several hundred tokens, yet many real cases are short. We study score-based detection as a function of passage length. We give a closed form for how the false-positive rate of a threshold set on long text changes as length falls, and a formula for the number of tokens a detector needs to reach a target true-positive rate. We calibrate thresholds within length bins with conformal p-values, and we add a bin-level rule that abstains when the detector cannot be informative at that length. On synthetic scores the formulas agree with simulation, and binned calibration holds the false-positive rate at its nominal value at every length. On three zero-shot detectors (log-likelihood, log-rank, Binoculars), three corpora and two open-weight generators, binned calibration gives a mean false-positive rate of 5.0% in every detector-length cell from 10 to 200 tokens, averaged over 300 random document-level resplits. A single threshold set at 200 tokens gives 29% for Binoculars at 10 tokens and about 1% for log-likelihood, as the closed form predicts. No detector reaches an 80% true-positive rate, and the best, Binoculars at 200 tokens, reaches 59%, so the rule abstains in every cell at that floor and releases only the 200-token bins at floors of 0.3 and 0.5.

arXiv AI Papershace 5 días

Robust Satisficing Ensemble: Scalable Model Aggregation Under Distribution Shifts

Standard ensemble methods can improve robustness by mitigating individual model stochasticity, but they frequently fail to generalize beyond the training distribution, leaving systems vulnerable to out-of-distribution (OOD) data. We introduce the Robust Satisficing Ensemble (RSE), a scalable probabilistic aggregation framework designed to secure model combinations by minimizing their fragility against distributional changes. By bridging a robust satisficing objective with a quadratic risk surrogate, RSE derives an analytical inner adversary. This fundamental reformulation enables highly efficient weight updates without the computational bottleneck of standard minimax optimization, crucially decoupling optimization complexity from sample size to scale effectively to large datasets. In particular, we prove a geometrically decaying truncation error for the fragility approximation via Lagrange-Burmann expansion and establish stationarity guarantees for RSE. We evaluate RSE across a comprehensive spectrum of distribution shifts, including adversarial label shift (SST-5, TREC), natural subpopulation and temporal shifts (CivilComments, HuffPost). Our results demonstrate that RSE consistently outperforms strong robust baselines, including GroupDRO, IRM, and SRM. These findings establish RSE as a practical, theoretically grounded safeguard for reliable model aggregation in dynamic environments.

arXiv AI Papershace 5 días

Sampling-based Model Predictive Control as Sequential Monte Carlo Inference

Sampling-based model predictive control (MPC) is treated as a family of distinct algorithms: predictive sampling, MPPI, the cross-entropy method, and their biased, colored-noise, and annealed variants all sample control sequences and reweight them by cost. We show a single Sequential Monte Carlo (SMC) inference algorithm implements them all, specified by three independent axes: the proposal, the update rule that reduces cost-weighted particles to a nominal plan, and the effective-sample-size threshold for resampling. We evaluate SMC-based planning on MuJoCo non-linear control, open-loop navigation through obstacle fields, 7-DoF motion planning, and non-prehensile push-T, in simulation and on a Franka FR3. Resampling combats particle degeneracy at long horizons. At a 28-knot horizon in our navigation domain it cut the collision rate by 30% over non-resampling baselines. Programmable proposals improve push-T success over perturbative baselines, improving from 1% to 100% in simulation and 10% to 90% on hardware.

arXiv AI Papershace 5 días

BRAT: Bounded Residual-Aligned Training for Stable Multi-Pass LLM Post-Training

Reinforcement learning for large language model post-training often reuses each batch of sampled rollouts for several gradient passes to amortize the cost of generation. Under such multi-pass reuse, existing objectives fail in different ways: unclipped updates collapse as the policy drifts away from the sampling policy, adding a negative baseline avoids collapse but reinforces every sampled response, and PPO-style ratio clipping cuts off gradients abruptly. We first ask why reuse is fragile. Measuring how well an update estimated from a small rollout group aligns with the KL-regularized target on held-out rollouts of the same prompts, we find that this alignment drops by nearly an order of magnitude from in-sample to held-out rollouts across three base models, suggesting that repeated passes largely overfit the sampled group. Motivated by this finding, we propose Bounded Residual-Aligned Training (BRAT), motivated by a proximal update toward a bounded reweighting of the sampling policy. BRAT assigns each token a gradient weight that decreases linearly with its probability ratio and vanishes at a trust-region boundary, and normalizes group-centered rewards by their maximum magnitude, keeping advantages zero-mean and bounded without extra hyperparameters. With four reuse passes, BRAT avoids training collapse, limits policy drift, and achieves higher average accuracy with lower seed variance on most benchmarks.

arXiv AI Papershace 5 días

The Surprising Effectiveness of Deleting Weights in LLM Post-Training

How little of a pretrained large language model (LLM) has to change for it to acquire new capabilities? We find that zeroing fewer than 0.05% of its weights, with no other modification, matches full fine-tuning (FFT) at scale across three LLM post-training regimes: reinforcement learning with verifiable rewards, supervised fine-tuning, and on-policy distillation. Our method, Deletion Tuning, learns a binary mask over the gate projection of transformer blocks. Two practical advantages follow: the trained mask is hundreds to thousands of times smaller than the corresponding FFT checkpoint, and Deletion Tuning forgets far less, which addresses a central challenge in continual learning. Under reinforcement learning, its prior-task accuracy stays near the base level while FFT, LoRA, and random sparse fine-tuning lose up to 5.4 points. To understand why masking alone suffices, we analyze the geometry of the learned weight changes under reinforcement learning using stable rank, spectrum drift, and principal-weight overlap. Compared with LoRA and random sparse fine-tuning, Deletion Tuning is closest to FFT across the three measures.

arXiv AI Papershace 5 días

SWXAI: A Unified Multimodal Explainability Library for Space Weather Forecasting and Heliophysics

Space weather forecasting and heliophysics increasingly rely on machine learning models that process heterogeneous inputs such as multi-viewpoint multi-wavelength solar images and rasters, multivariate time series, spatial annotations, and tabular event-level data. However, explainability workflows remain fragmented across libraries, model interfaces, and output formats, limiting reuse and reproducibility. We present SWXAI, an open-source library that unifies explainable AI (XAI) workflows for image, time-series, and tabular data in space weather forecasting and heliophysics. SWXAI provides a common, extensible interface across commonly used vision, time-series, and tabular learning models, with validation, adaptation, and persistent explanation objects. It also includes domain-grounded examples, reproducible notebooks, tests, and documentation, enabling users to inspect outputs, swap models or explainers, and extend workflows to new heliophysics datasets. SWXAI reduces modality-specific engineering overhead and supports comparable, reusable explanations across heterogeneous scientific machine learning pipelines. A benchmark against direct backend implementations shows that SWXAI substantially reduces the code users write to produce an explanation, with no practical runtime cost.

arXiv AI Papershace 5 días

Structurally Separated DAG Learning with Multi-Scale Normalized Closure

Continuous DAG learning commonly enforces acyclicity through scalar functions of a weighted adjacency that aggregate evidence from weighted closed walks. We add a complementary layer: matrix-valued closure logits---explicit multi-hop structural relation scores for each ordered pair---supervised alongside data fit, sparsity, and a DAGMA-style barrier on the weighted adjacency. The distinction is not scalarization of the overall objective but how path evidence is represented and exposed before aggregation. SMNC (scale-normalized multi-scale closure) builds these scores on a structural gate by repeated squaring in log space and divides each scale by its path-length budget, counteracting product-style attenuation along long cycles; diagonal entries support node-wise cycle supervision and off-diagonals optional pairwise reachability cues. In the linear model, a bounded structure--magnitude factorization ties strong effective edges to strong gates; a nonlinear induced gate offers another construction with the same intent. For sufficiently strong cycles, normalized logits admit a length-independent lower bound; under a unique-active-cycle assumption, the hard final-scale readout yields a polynomial log-gate gradient factor rather than exponential product-style decay. On cue-free, unstandardized Gaussian linear and nonlinear ER benchmarks, these models improve over DAGMA, run with default hyperparameters at a reduced budget, on denser graphs and are close to it on sparse ones.

arXiv AI Papershace 5 días

A Geometry-Guided Feature-Space Framework for Interpretable Unsupervised 4D-STEM Analysis of Complex Multiphase Materials

Rapid, high-fidelity analysis of four-dimensional scanning transmission electron microscopy (4D-STEM) datasets from heterogeneous materials remains challenging because coexisting phases, orientation variants, local strain gradients, and intrinsic TEM thin-foil effects produce complex, overlapping diffraction signatures. Here, we present a machine-learning (ML) framework that combines physics-informed geometric descriptors with weighted high-dimensional embeddings to identify latent structural patterns with minimal prior knowledge. By constructing a regularized descriptor space before embedding and density-based clustering, the method improves cluster stability, suppresses noise-driven fragmentation, and reduces sensitivity to embedding hyperparameters. Spatial weighting produced the largest stability gain, increasing median cluster persistence from 0.04 to 0.55 while reducing the noise fraction from approximately 22% to 0%. Rotation weighting increased DBCV from 0.80 to 0.92, although the noise fraction rose from 0.2% to 1.0%, highlighting a trade-off between improved density separation and noise rejection. We validate the framework on two benchmark systems: Si/Si_xGe_{1-x} epitaxial layers with continuous lattice variation and deformation, and multiphase Ti-Nb alloys containing distinct phases and orientation variants. Comparisons with a pretrained vanilla autoencoder and model-free pattern vectorization show that the physics-informed geometric descriptor provides higher cluster fidelity, greater interpretability, and substantially lower dimensionality.

arXiv AI Papershace 5 días

Inference for Reasoning Models: Practical Lessons from Evaluation and Serving

Running a language model involves more than its trained weights. The same model can receive very different benchmark scores, produce different answers, or cost more to serve depending on how it is evaluated and run. In one experiment in this book, changing only the maximum response length moved accuracy on a math benchmark from 0.08% to 93.18%. This book explains how inference settings shape the results we report and the systems we deploy. We begin with evaluation, including generation limits, answer extraction, scoring methods, and quantization. We then examine why outputs can change across batches, attention backends, and caching, even when sampling is disabled. The next chapters cover serving performance, including batching, memory capacity, latency, throughput, and cost. We also study structured outputs, tool calls, reasoning budgets, and long-context tasks, showing what these controls and benchmarks do and do not establish. The book concludes with practical guidance for recording inference runs, checking evaluations, estimating capacity, and comparing systems fairly. Written for researchers and engineers working with language models, it aims to make inference results easier to understand, reproduce, and trust.

arXiv AI Papershace 5 días

Factor Rank: A Canonical Algebraic Generalization of Tree Decision Diagrams

Tree decision diagrams (TDDs), recently introduced by Capelli et al. (SAT 2026), are representations of Boolean functions. They generalize OBDDs by replacing variable orders with vtrees. The width of a TDD counts the distinct subfunctions across each vtree cut. Counting subfunctions is blind to linear structure. We introduce algebraic tree decision diagrams (ATDDs) by replacing this count with the rank of the cut matrix. The maximum of these ranks over all cuts, the factor rank, is exactly the width of the ATDD. Every TDD can be converted into an ATDD of no greater width. An ATDD is a tree tensor network in canonical form. ATDDs are exponentially more succinct than TDDs, and quasi-polynomially more succinct than structured deterministic DNNFs. Our main result is that this gain in succinctness does not cost canonicity. ATDDs are the first strict canonical form for a rank-based representation over arbitrary vtrees. For each function and vtree there is exactly one ATDD, over the rationals and without any genericity assumption. The canonicity admits a local characterization. Whether a tree tensor network is the ATDD of the Boolean function it computes can be checked in polynomial time from its tensors alone. Equivalence checking becomes a test of syntactic equality. The operation set includes polynomial-time model counting, weighted model counting, and a test whether a tree tensor network represents a Boolean function. Capelli et al. asked for the complexity of changing the vtree of a TDD and suggested local vtree changes. We answer the question. After a rotation of the vtree at one node, the new ATDD differs from the old one in two tensors only. Vtree change runs in output-polynomial time both for TDDs and for ATDDs. We formalize and verify our core results in Lean. We provide a prototype implementation of TDD and ATDD compilers as a first empirical validation.

The Verge AIhace 5 días

AI broke the cybersecurity world. What’s next?

Today, I’m talking with Alex Stamos, chief security officer at Cognition and longtime cybersecurity expert, about what a mess cybersecurity is right now. This episode…

Hacker News (AI)hace 5 días

AARE: A proof-of-concept open-source implementation of Android Auto

en▲ 5

arXiv AI Papershace 6 días

Sampling Is Not Curiosity: Why LLM Agents Should Investigate

LLM agents are increasingly used for open-ended tasks, very common in research scenarios where they are required to explore, investigate, or act curiously. At inference time, an LLM is a fixed map from context to next-token distribution, the pretrained weights determine the function, and only its conditioning changes as context accumulates. Mechanisms like temperature, best-of-N, tree search, or RAG vary how the distribution is sampled or what it is conditioned on, but do not update the agent's model of the environment. We argue that current LLM agents do not explore curiously; they explore stochastically. From how curiosity is understood in the RL literature, it requires a predictive structure over how the environment responds to the agent's actions, which improves online from current interaction, drives action selection through its predicted improvement, and persists beyond the context window. Additionally, in its open-ended form, the agent should intrinsically originate what to investigate. Thus, we propose a framework defining curiosity in LLM agents, reserving the term for systems that satisfy it. In doing so, we identify the research directions its implementation opens up.

arXiv AI Papershace 6 días

Safer for Whom? A Systematic Exploration of AI Safety in Eating Disorder Interactions

AI assistants advise users about diet, exercise, and weight loss---areas in which unsafe guidance can worsen body dissatisfaction, dietary restriction, compulsive exercise, and other behaviors associated with eating disorders (EDs). Existing evaluations emphasize diagnostic accuracy, overtly harmful prompts, or synthetic scenarios, providing limited insight into whether models recognize implicit vulnerability in natural interactions or protect users consistently. We evaluate responses from three open-weight large language models to 100 questions sourced from an online pro-anorexia community. We grade responses using clinically informed criteria that extend a seven-dimensional taxonomy of ED-related harms. We also construct matched conditions in which questions are accompanied by information about the user varying race, gender, body mass index (BMI), and explicit ED vulnerability. Safety varies substantially across models and harm dimensions. High-risk responses are approximately twice as frequent without bios, suggesting that models do not consistently recognize vulnerability from language alone. Explicit signals of ED vulnerability improve safety by 10--14 points across models. Responses also differ significantly by race, gender, and BMI, with body size producing the largest disparities. These patterns are consistent with reliance on the stereotype that EDs primarily affect thin, White women, potentially leaving users outside this profile with less protective responses.

arXiv AI Papershace 6 días

GEAR: A GPU-Accelerated Global Solver for Bounded Continuous Nonlinear Programs via Linear Bound Propagation

We present a novel GPU-accelerated global solver for constrained nonlinear programming (NLP) inspired by the efficient linear bound propagation framework in neural network (NN) verification. Existing work on NN verification, a problem that can be cast as feasibility detection for a nonlinear objective, demonstrated orders-of-magnitude speedups with the linear bound propagation-based framework when compared to traditional solvers. Extending this paradigm to global nonlinear programming requires addressing many challenges: a global constrained NLP solver must effectively exploit all, possibly unstructured, nonlinear constraints to improve the objective, whereas NN verification typically handles simple input box constraints over a well-structured NN and does not directly optimize an objective function. We address these challenges by fully exploiting the linear bounds produced by bound propagation: these linear bounds can be viewed as relaxations of nonlinear constraints and used to provide lower bounds, check infeasibility, and guide the search for a primal solution. In addition, we propose tighter linear relaxations for bilinear, quotient, and other commonly used nonlinear functions in NLP problems, and a projected augmented Lagrangian method tightly coupled to our dual solving procedure to produce primal solutions. Our method solves 421 of 505 bounded and continuous NLP problems in the GAMS and MINLPLib benchmarks under a 180-second limit, more than each of the six global solvers we compare with, including Gurobi and BARON (commercial), and SCIP (open-source). Sixteen of these problems are certified by none of the six global solvers; all of them contain exponential, logarithmic, or trigonometric terms, which are often challenging for traditional approaches. Project webpage is at https://verified-intelligence.github.io/GEAR.