2440
178
Tata Consultancy Services (TCS) built an agentic test orchestrator on Amazon Bedrock AgentCore that turns requirements into executable, governed tests for its Banking, Financial Services, and Insurance (BFSI) clients. Learn how specialized agents, per-tenant isolation, and workload-aware execution cut testing effort by 88% in an internal benchmark.
Running Amazon Bedrock across many accounts shows you the bill but not what happened inside each call. This post builds a centralized observability data lake that aggregates Bedrock invocation logs from every linked account into one queryable table, so you can see prompts, token counts, model choice, and per-invocation cost across your organization.
Emergencies can occur where there is no phone signal and no ambulance nearby. A small language model on a phone can be helpful in such situations, but its advice needs to be safe and accurate. We built HandsOnAid, a 4-bit Gemma 2B model fine-tuned with QLoRA on FirstAidQA dataset, a corpus of 5,550 first-aid question-answer pairs. We compared seven configurations, including the base, 4-bit and 8-bit fine-tuned models, and versions with a length floor, safety gate, or retrieval. We used a 41-pair question-answer bank with medically reviewed references that followed ANZCOR guidelines, and a strict 0-5 score rubric, with twelve safety overrides. Four frontier LLM judges (GPT-5.6 Sol, Claude Opus 5, DeepSeek V4 Pro, GLM-5.3) blindly scored every answer, with 45 hidden controls used to check judge reliability. Fine-tuning improved the mean score by +1.08 over the base model to 2.34. The best configuration scored 2.51 out of 5 which was the fine-tuned 4-bit model with retrieval. A medical professional also conducted a blind evaluation of 40 answers across the base, the 4-bit fine-tuned, and the retrieval configurations, and the scores were similar to the LLM judges' average scores. A keyword audit found limited coverage of several details of ANZCOR guidelines in the training corpus, that are required by the rubric, which suggests that missing information in the training data could be a cause of the low scores. On a smartphone device (Pixel 7 Pro) with six CPU threads and no network connection, the 4-bit fine-tuned model produced its first displayable text in 6.86 seconds and a complete answer in 14.88 seconds on average over ten questions.
Adversarial training is often plagued by "pseudo-attacks", perturbations that significantly increase the loss value but fail to truly alter the model's confidence structure. To address this, we propose two complementary approaches to directly attack and defend against confidence structure deformation. First, building upon the traditional loss maximization perspective, we introduce a hybrid objective that integrates cross-entropy and KL divergence, encouraging adversarial examples to simultaneously disrupt prediction accuracy and distribution consistency. Second, we geometrically reconstruct adversarial training by generating perturbations that maximize the distribution distance on the confidence manifold. Theoretical analysis shows that the optimal perturbation direction aligns with the principal eigenvectors of the Fisher information matrix. We also propose an efficient algorithm to avoid high-dimensional matrix computations. Experiments on multiple datasets and model architectures demonstrate that our method outperforms existing methods in adversarial robustness and robust transferability, particularly excelling in structural perturbations within high-confidence regions.
Foundation models pre-trained on large unlabeled electroencephalography (EEG) corpora have emerged as a transformative paradigm for Brain-Computer Interface (BCI) applications. Recent EEG foundation models have shown that multi-scale representations derived from fixed-length patches can improve EEG representation learning. Existing frameworks commonly construct temporal scales through feature-level operations. However, these operations leave the correspondence between learned features and scale-specific characteristics of the original EEG signals implicit. Moreover, when the self-supervised objective is not aligned with the multi-scale input representations, information at some temporal scales may receive insufficient supervision during pre-training. In light of this, we propose ScaleBrain, a scale-consistent pre-training framework that explicitly coordinates multi-scale representations of EEG from both the model input and optimization objective sides. Specifically, ScaleBrain introduces multi-scale EEG tokenization to represent the same EEG segment through analysis windows with different temporal spans and shared temporal anchors, and pairs these representations with scale-specific reconstruction to align with the multi-scale input structure and provide supervision across temporal resolutions during self-supervised learning. We further develop an axial spatio-temporal attention mechanism to contextualize the fused multi-scale representation along spatial and temporal dimensions. Extensive evaluations on six public BCI benchmarks against 12 representative baselines show that ScaleBrain achieves the highest mean Balanced Accuracy on five datasets and ranks third on the remaining one, with only 5.0M parameters, demonstrating strong performance and favorable parameter efficiency.
Long-horizon personalized agents maintain an evolving memory by continuously deciding what to add, modify, or discard. However, learning an optimal update policy remains challenging. Conventional training relies solely on terminal question-answering accuracy, yielding a coarse feedback signal that treats all intermediate updates identically. Furthermore, update utility is inherently user-specific, as identical memory actions may yield opposite effects across different user personas. Existing methods fail to address this heterogeneity, relying on globally shared evaluation standards with at most a single scalar intermediate reward. To address these challenges, we propose Personalized Memory Rubrics (PMR), a framework that translates terminal feedback into fine-grained, per-step, and per-user credit assignment. PMR decomposes evaluation into a universal layer capturing cross-persona failure patterns and a per-user adaptive layer modeling individual traits as a specialized delta over the universal baseline. Through Rubric-Guided Credit Assignment, PMR redistributes credit at the token level, selectively amplifying validated decisions on successful trajectories while falling back to standard policy optimization on failed ones to avoid unreliable attribution. The rubric and memory policy co-evolve in a closed loop with anti-circularity guarantees. Crucially, PMR incurs zero inference overhead, operating without any rubric or external judge during deployment. Extensive experiments across four benchmarks demonstrate consistent gains over strong memory-augmented baselines.
In reinforcement learning with verifiable rewards, a language model is rewarded for a correct final answer, and GRPO passes that reward to every token of the solution in equal measure. We instead place the credit where the model hesitated, in a method we call MCPO (Margin-Credit Policy Optimization). We measure hesitation by the top-2 margin M=p_1-p_2 of the policy's next-token distribution, map it to a score with two fixed thresholds, and spread each solution's advantage over its tokens so that the mean stays unchanged. The margin is read from a forward pass the trainer already performs, so the method needs no extra model evaluation and no batch statistics. We train MiniCPM5-2B-Base with LoRA for 200 steps on 2,048 DAPO-Math problems and evaluate on six competition-math benchmarks. MCPO beats an unmodified GRPO baseline for all five data seeds, by 6.55 points of macro avg@32 on average (95\% CI [4.1, 9.0]), and a Qwen3-4B-Base replication improves for all three seeds (+4.75). The gain is largest on problems that GRPO solves only some of the time and grows with k in pass@k, reaching 13.8 points at pass@32. MCPO models also write two to three times longer answers, and runs that write longer score higher; our experiments do not separate the two effects.
Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. They must therefore decide what to write down and what to read back, and when to do either. Recent work has begun to use on-policy distillation (OPD) to make memory control part of the policy, relying on memory tools and compaction protocols that lie outside either model's pre-training. However, this weakens the distillation on both sides: the teacher cannot give useful guidance on an interface it has never used, and the student samples too few meaningful memory actions for that guidance to take effect. To study this problem, we introduce Coding Agent Memory Gym II (CAMG-II), a suite of long-horizon agentic-RL environments spanning Shop, Coding, AutoResearch, General, and Visual. Alongside each environment's native task interface, CAMG-II provides ordinary Bash access and an episode-persistent workspace, enabling the policy to create, revise, search, and reuse files as memory across context replacements. We also introduce Coding Agent Memory Post-Training II (CAMP-II), which trains a single policy jointly across all five environments with fully asynchronous OPD to learn this file-based memory behavior directly from a frozen teacher with a prior for file operations. On CAMG-II, CAMP-II obtains 2.97 times the gain of the strongest OPD baseline. The memory behavior CAMP-II learns generalizes to two external benchmarks, Terminal-Bench 2.1 and SWE-Bench Pro V2, with improved success rates.
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities but raise significant privacy concerns due to their abilities to infer sensitive personal information from images with high precision. While current LVLMs are relatively well aligned to protect universal privacy, e.g., credit card data, we argue that privacy is inherently personalized and context-dependent. This work pivots towards a novel task: can LVLMs achieve Inference-Time Personalized Privacy Protection (ITP^3), allowing users to dynamically specify privacy boundaries through language specifications? To this end, we present SPY-Bench, the first systematic assessment of ITP^3 ability, which comprises (1) 32,700 unique samples with image-question pairs and personalized privacy instructions across 67 categories and 24 real-world scenarios, and (2) novel metrics grounded in user specifications and context awareness. Benchmarking the ITP^3 ability of 21 SOTA LVLMs, we reveal that: (i) most models, even the top-performing o4-mini, perform poorly, with only 24% compliance accuracy; (ii) they show quite limited contextual privacy understanding capability. Therefore, we implemented initial ITP^3 alignment methods, including a novel Noise Contrastive Alignment variant which achieves 96.88% accuracy while maintaining reasonable general performance. These results mark an initial step towards the ethical deployment of more controllable LVLMs. Code and data are available at https://github.com/achernarwang/specify-privacy-yourself
Reinforcement learning from formal specifications often separates task definition, reward shaping, curriculum design, and policy diagnostics into different mathematical objects. This fragmentation requires additional conversion steps and discards structure already present in the specification, such as subtask boundaries, ordering constraints, and quantitative measures. We introduce TBTCL (Temporal Behavior Tree Centric Learning), a framework that uses Temporal Behavior Trees (TBTs) as a unified interface for specification-guided reinforcement learning. A single TBT specifies the task, induces dense robustness-based rewards, defines a curriculum, and supports better understanding of learned policies. By adapting TBT syntax and semantics for specification-guided reinforcement learning, we provide a formalism that is more expressive than popular specification languages such as LTL and inherently hierarchical. We present algorithms to automatically extract a curriculum DAG and derive dense, robustness-based rewards directly from TBT semantics. Furthermore, we utilize TBT trace segmentation to build understandable competence profiles and automatically adapt the reward toward bottlenecks. We evaluate TBTCL on six specifications across discrete and continuous domains and show that Temporal Behavior Trees can serve as a specification, a reward, a curriculum, and a diagnostic.
Taxonomies are structured knowledge representations that support semantic search, recommendation, product catalog organization, and retrieval-augmented generation. Their automated construction from a fixed set of candidate entities requires more than plausible text generation: the resulting hierarchy must be complete, structurally valid, and semantically coherent. Large Language Models (LLMs) are promising for this task, but iterative free-form generation remains vulnerable to omitted or hallucinated entities, invalid hierarchical relations, and inconsistent organizational principles. We introduce TIGA (Taxonomy Induction with Tool-Grounded Agents), a knowledge-engineering architecture that formulates taxonomy induction as an iterative decision process over an explicit state and constrained action space. TIGA combines executable taxonomy-editing tools, an memory that records tool calls, rationales, and outcomes, and a runtime-adaptive organizational policy that the agent revises during inference. An optional global reviewer evaluates the constructed taxonomy against the candidate set and the active policy. We evaluate TIGA on four benchmark taxonomies using two LLM backbones and compare it with single-shot and iterative prompting-based baselines. TIGA achieves the strongest structural-quality results in most evaluated configurations, but does not dominate on every precision measure. The gains are accompanied by substantially higher token consumption, API-call volume, and inference time. Ablation results show consistent benefits from the policy and interaction-memory components, whereas the effect of global review depends on the dataset and the available revision budget. These findings characterize tool grounding and runtime policy adaptation as promising but costly mechanisms for improving the controlled construction of structured knowledge representations.
AI coding harnesses couple language models to shells, filesystems, and network sockets, turning the harness into an automated Living-off-the-Land (LotL) platform: once an adversary injects instructions into retrieved content, dual-use binaries such as bash, curl, and find become the payload. We argue that security in agentic systems is a structural property of the cognitive architecture and permission composition rather than an emergent property of individual models. Static regex blacklists miss structural evasions, whereas LLM-as-a-Judge guardrails incur high latency while over-blocking benign work. We present ActiveSecHarness, a neurosymbolic reference-monitor architecture built around Separation of Intelligence and Authority. It combines (1) Macaroon-inspired HMAC capability tokens across four Trust and Authority Rings, (2) a 2D product taint lattice {L} = {L}_{trust} {L}_{sens} and a workspace provenance ledger that prevents secret laundering across disk, (3) deterministic Ring-0 OS/tool guards, and (4) bytecode certificates, split-horizon audit logs, self-healing security test synthesis, and an adaptive UserWikiManager. Across N=252 fixtures (120 attack, 120 benign, 12 kill chains), a live local LLM judge (qwen3:4b) refuses 98.3\% of benign work at 142.3\,ms latency, whereas Ring 0 achieves ABR=1.000 and FPR=0.000 in 12.7\,μs (11{,}200faster). In a multi-step trajectory ablation, an oracle Conseca-style stateless policy misses 4/12 kill chains (ABR=0.667) and coarse Boolean taint blocks 83.3\% of benign workflows, while our 2D product lattice achieves ABR=1.000 and FPR=0.000 at 20\,μs per step.
} Predicting who will speak next, and when, is a prerequisite for conversational agents and social robots that take turns without waiting for silence. Multiparty next-speaker prediction (NSP) has been studied on corpora with per-speaker sensors, each with its own event definitions. No public benchmark covers the single camera and microphone a robot has, and no model has been compared against how well people perform the same prediction. We turn AVCC, a 31-hour audio-visual corpus of two- and three-speaker online conversation recorded with a single static camera and microphone, into a public NSP benchmark. We add a speaker-disjoint seven-video test partition with 865 hold and 865 shift events at mutual-silence gaps, word-level forced-aligned transcripts for the whole corpus, and a human reference benchmark collected at the same events. We score an audio-only Role-Relative VAP model (RoleVAP) and the multimodal MuVAP model. Labels, transcripts, per-response human data, and the scripts that reproduce every number are released.
Latent diffusion models decode approximately correct latents, making decoder sensitivity to perturbations relevant to image quality. After whitening, forward diffusion corrupts every direction at the same signal-to-noise ratio (SNR), despite unequal decoder sensitivities. We study an invertible linear layer between the encoder and diffusion process that shifts each direction's log-SNR. Its linear minimum mean-square-error (LMMSE) recovery risk has an exact spectral expression for any whitened latent distribution. Under an energy budget and a floor ensuring invertibility, minimizing decoder-sensitivity-weighted LMMSE risk integrated over the diffusion schedule yields a unique optimal allocation. The layer aligns with the decoder's expected pullback metric and equalizes marginal risk reduction across directions, computed by a deterministic spectral solver. At large budgets, the allocation follows a power law in sensitivity whose exponent is determined by the low-SNR tail of the loss weighting: one half for noise prediction and two thirds for data prediction under the cosine schedule, with a crossover on truncated schedules. We instantiate this principle as RB-LDC using Jacobian-based sensitivity estimates. On ImageNet at 256 x 256 resolution with SD-VAE and VA-VAE, trained RB-LDC improves reconstruction from noisy latents over budget-matched isotropic coding in all four metrics and five coding configurations, averaged over 15 SNRs and three pipeline seeds. It reduces average perturbed FID by 17.1 for VA-VAE block-local coding while preserving clean reconstruction. We also report intermediate SNRs where perturbed FID worsens. With a frozen diffusion model, RB-LDC yields lower FID point estimates in all 12 tested combinations of sampling progress and channel SNR under corruption during sampling.
On bounded Euclidean domains, we study how hidden-parameter geometry affects shallow ReLU^k networks through results on approximation rates, saturation-type lower bounds, Gram and stiffness matrix spectra, Bernstein and Marcinkiewicz--Zygmund inequalities, discrete least-squares recovery, and inverse approximation. Our analysis uses smooth localization and dyadic Fourier estimates. We quantify parameter geometry by the fill distance h and antipodal separation distance h. For the prescribed network space L_n^k, representative approximation estimates take the form {equation*} {aligned} _{g_nL_n^k}\|f-g_n\|_{L^2(Ω')}&h^r\|f\|_{H^r(Ω)},\\ _{g_nL_n^k}\|f-g_n\|_{L^2(Ω)}&h^{{d+2k+1}{2}}\|f\|_{L^2(Ω')}. {aligned} {equation*} The upper bound exploits target regularity up to the critical order {d+2k+1}{2}, while the lower bound applies asymptotically to fixed targets smoother than this order. The upper estimate holds on an interior subdomain Ω', whereas the lower estimate concerns the larger domain Ω. We also obtain geometry-dependent spectral estimates that quantify conditioning and coefficient stability. Further results establish norm discretization and regularized recovery from function values, and infer target regularity from approximation rates under coefficient or separation conditions. Altogether, these results describe how parameter geometry governs approximation accuracy, the stability of coefficient representations, discrete recovery, and the regularity information contained in convergence rates.
Post-placement timing and power co-optimization is critical to meeting setup timing targets and reducing power within placement constraints. A cell replacement can improve timing along one path while increasing power or reducing slack on others. Conventional optimizers follow fixed rules for scheduling and accepting repairs, limiting their ability to address design-specific bottlenecks. A large language model (LLM) can propose changes to these rules, but on its own it cannot reliably estimate how a cell replacement affects endpoint slack and power. TAPCO-G combines LLM-guided optimizer adaptation with graph-based physical priors to evaluate candidate cell replacements. The outer loop tunes tool settings using physical-design feedback and revises executable repair policies when tuning stalls. The inner loop anchors delay estimates to static timing analysis snapshots, accounts for upstream driver-load changes, and propagates delay shifts to timing endpoints. It combines these estimates with coarse-to-fine power screening to prioritize candidates for physical trials. TAPCO-G produced placement-legal, logic-equivalent implementations and outscored CERBERUS on all eight ASAP7 benchmarks from the MLCAD 2026 Contest, with a mean quality score gain of 5.516 points, and scored 21.648 points higher than TAPCO on average. These results demonstrate the effectiveness of the proposed co-optimization framework.
Explainable AI (XAI) has expanded into a broad and impactful research area, but the fundamental question of how best to evaluate explanation techniques has proven frustratingly elusive. Methodologies typically revolve around either proxy metrics which tell us little about the utility of an explanation, or human evaluation which---while generally considered the gold standard---is expensive, difficult to scale, and challenging to reproduce. In this paper, we investigate if Multimodal Large Language Models (MLLMs) can act as surrogate human participants to automate the evaluation of XAI. To do this, we first propose a ``purpose-grounded'' evaluation framework with metrics applicable to both MLLMs and human users alike. Then, we offer some conceptual insights as to why and when we can trust these models to automate this process. Lastly, we conduct extensive empirical testing across five purpose-grounded areas such as model improvement and fourteen popular MLLMs. Results indicate that all models trend positively, but OpenAI and Qwen models generally show higher alignment compared to other variants, which would be the recommended choices in practice. We expect our insights not to replace user testing, but rather to help XAI researchers complement their current evaluation methodologies and move the field away from heavy reliance on proxy metrics for measuring explanation utility.
Automated fruit grading, sorting, and orientation systems require real-time, accurate detection of fruit-level morphological landmarks rather than the fruit boundary alone. This paper presents a custom, fully self-collected and self-annotated dataset of 1506 apple images (1054 training, 301 validation, 151 test) captured under controlled, fixed-camera conditions and annotated for three target classes: fruit body, stem cavity, and calyx. A YOLO11s detector trained with default hyperparameters is adopted as the baseline, and a systematic, one-factor-at-a-time ablation study is conducted across input resolution, data augmentation policy, optimizer configuration, and lightweight architectural enhancement to identify the factors that most improve detection of these fine-grained landmarks under high inter-cultivar color and texture variability. Neither increased input resolution, nor any of six tested augmentation variants, nor manual optimizer tuning improved on the untuned baseline (F1 = 0.9047); only a same-slot architectural substitution of the backbone's C2PSA block with Coordinate Attention (CA) matched the baseline's overall F1 (0.9038) while reducing parameter count by 10.2% and achieving the study's best [email protected] (0.935) and best stem-cavity F1 (0.8217). A confidence-threshold sweep identified 0.15 as the deployment operating point for this model, and a confirmatory experiment combining the CA-swap architecture with the best-performing augmentation variant showed the two effects do not combine additively, reversing the architectural gain rather than compounding it. The resulting design recommendations --- the CA-swap architecture alone, at a 0.15 confidence threshold --- target real-time apple landmark detection for orientation-aware industrial sorting and alignment systems.
Open-Vocabulary Scene Graph Generation (OV-SGG) requires models to recognize visual relationships beyond the training vocabulary, yet existing methods often rely on dataset-specific object--predicate co-occurrence patterns. Under long-tailed distributions and noisy object localization, such reliance leads to biased, entangled relation representations, limiting both rare-relation recognition and generalization to unseen predicates. To address these challenges, we propose RaRe, a reasoning-aware relational representation learning framework that transforms MLLM-generated chain-of-thought (CoT) rationales into structured relation embeddings. Rather than using CoT only as an intermediate reasoning trace for final prediction, RaRe directly optimizes the rationale-derived representation space to improve inter-relation discriminability. Specifically, RaRe first aligns rationale embeddings with textualized relation descriptions through self-supervised contrastive learning, and then refines rationale generation with a specificity-oriented reinforcement learning objective that encourages semantically object-pair-specific relational evidence. Experiments on VG150 and PSG demonstrate improved category-balanced recognition over existing baselines, while the additional gains from learning vary across datasets and proposal settings. Code available at: https://artanic30.github.io/project_pages/RaRe.
