August 26, 2026

Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers

Multi-vector embedding models, also known as late-interaction or ColBERT-style architectures, avoid compressing an entire text into one vector. Instead they retain a contextualized embedding per token and score queries against documents with the MaxSim operator, which sums the best token-level matches.

August 26, 2026

Claude Opus 4.6 Bypasses Gym Booking Limit, Cancels Other Users' Reservations in Tests

Aikido Security recreated the Australian gym-booking incident inside a controlled synthetic environment to examine how an autonomous agent behaves when given access to a flawed API. Using Claude Opus 4.6 on the OpenClaw v2026.4.1 harness, the model was instructed only to secure consistent gym-class bookings.

August 26, 2026

Fake Apple Support AI Calls Target Stolen-Device Owners for Passcodes and 2FA Codes

Cybersecurity researchers have disclosed details of a phishing-as-a-service (PhaaS) platform built to strip Apple's Activation Lock from stolen devices, using rented AI voice agents that call theft victims posing as Apple Support and ask for their device passcode. SOCRadar Threat Research Unit (STRU) said the platform, which it tracks as AnonyMousKIT, is credit-metered and drives lures across

Rel 85

August 26, 2026

Function-Level Execution Feedback for Code Preference Optimization

Process supervision has improved mathematical reasoning, where intermediate steps are naturally expressed as chains of thought. In code generation, however, process supervision remains underexplored because there is no standard notion of a step. Supervision can target lines, reasoning traces, or program states, making it unclear what to label and optimize. We propose STEP-KTODER, a framework for code preference optimization that defines steps as module-level functions in decomposed multi-function programs and assigns binary correctness labels via automatically generated unit tests. Our method provides a code-specific instantiation of stepwise KTO, combining function-level process supervision with outcome-level feedback on the full program. We evaluate on HumanEval(+), MBPP(+), BigCodeBench, and LiveCodeBench, showing that STEP-KTODER improves over outcome-only KTO and DPO. Further analysis shows that execution-based labels are essential: LLM-as-a-judge annotations systematically over-predict function failures, corrupt positive step labels, and degrade downstream preference optimization. Code is available at: https://github.com/inechnech/STEP-KTODER.

Rel 85

August 26, 2026

LLM Agents Perform Controlled Experiments Using Simulation Models

Large language models (LLMs) have shown strong capabilities in reasoning, planning, and tool use, but many scientific and engineering tasks require more than plausible text and code generation. They require understanding how a system responds to intervention, which in practice depends on controlled experimentation. In this work, we propose a multi-agent framework that enables LLM agents to conduct controlled experiments with scientific simulation models for pharmaceutical process design. Given a user query and a baseline configuration, the system constructs a structured task representation, designs experiments, executes comparative simulation, interprets the resulting outcomes, and synthesizes evidence-based recommendations for process parameter optimization. By coupling language models with high-fidelity simulation models in an interactive agent framework, the proposed system supports reasoning through intervention, comparison, and observation. As a result, it produces more specific and actionable outputs than language-only reasoning. In an industrial application setting, this advantage is reflected in higher output specificity as well as improved user-rated correctness and helpfulness. Ablation studies and visualized case analyses further demonstrate the effectiveness and practical utility of simulation-integrated experimental reasoning.

Rel 85

August 26, 2026

Automata from Agent Traces: Failure and Next-Step Prediction

LLM-based agents execute multi-step tasks, but their behavioral structure remains opaque: long unstructured traces resist the safety auditing and runtime monitoring that deployment requires. Existing approaches operate per-trace or success-only, so they miss the cross-run topology that links next-step and failure prediction. To recover that shared structure, we collapse an entire trace corpus into a single, compact finite-state machine (FSM) that serves as a structural substrate for the otherwise unpredictable behavior of LLM agents. Across twelve public datasets, the FSMs are compact (7-43 states), replay held-out data at >=0.997 fitness with near-identical topology across splits, and build in milliseconds. This substrate addresses both prediction goals. For next-step prediction, FSM-state context outperforms Agent Workflow Memory on every ground-truth-matched dataset. For failure prediction, per-state behavioral features reach held-out AUROC up to 0.94, and an online monitor ranks failing runs above passing ones from a partial trace, triggering early stopping well before completion. Behavioral topology thus appears shaped more by the deployment harness than by the LLM, providing a model-agnostic structural primitive for safety auditing and runtime monitoring.

Rel 85

August 26, 2026

Serving Masked Diffusion LLMs: Characterization and Design Principles from Real Hardware

Masked diffusion language models (dLLMs) can in principle generate text faster than autoregressive (AR) models, since they denoise many tokens at once. Recent systems have begun building serving infrastructure for dLLMs, but none first measure how these models behave under real, concurrent serving load. Serving systems built without this grounding risk carrying over assumptions from AR serving that may not hold for dLLMs. We characterize dLLM serving to close this gap, using LLaDA-8B-Instruct with a D2F (Discrete Diffusion Forcing) LoRA adapter on a single NVIDIA H200 GPU, evaluated on GSM8K and HumanEval. We report three findings. First, request difficulty, the number of denoising steps a request needs, is discrete rather than continuous: requests fall into 11 fixed step-count levels (178 + 29k), and no signal we test predicts the level before generation starts (best R2 = 0.150). Second, benchmarks with short generation budgets below 320 tokens understate serving variance, since requests are cut off before the latency spread appears. Third, only 24% of single-request wall-clock time is GPU computation; the rest is CPU-side dispatch overhead. Batching mainly helps by amortizing this overhead: sharing one forward pass per denoising step improves throughput by 16.0x at batch size 16 over a per-request-dispatch baseline. We also argue structurally that output quality should not degrade with batch size, stating three assumptions this rests on; we measure 74 to 76% GSM8K accuracy at single-request scale. Finally, we derive a batch-timeout rule for fixed-fill synchronized batching under Poisson arrivals. Together, these results show that serving diffusion language models needs parallelism at the level of each denoising step, which differs from AR serving in how admission and eviction interact with an already shared forward pass.

Rel 85

August 26, 2026

Autonomous Mathematical Discovery in an Open-World Multi-Agent Environment

We study autonomous mathematical discovery in the Station, an open-world multi-agent environment in which AI agents from different model families pursue a shared research goal without a central coordinator or scripted pipeline. Agents choose their own research directions, conduct experiments, collaborate, and build a shared scientific literature. Across 12 construction problems from the AlphaEvolve catalogue and two additional case studies, the Station obtained results novel relative to the prior literature on five problems: a new infinite family of finite-field Kakeya sets, new exact 604-point kissing configurations in dimension 11, new records for the discretized Kakeya needle and sign uncertainty problems, and a substantially improved lower bound for Erd\H{o}s's minimum-overlap problem. Agents also discovered novel infinite families for Book Ramsey numbers. Importantly, the agents produced not only numerical constructions but also theorems and analyses explaining how those constructions work, making the results more interpretable and easier for mathematicians to build upon. We release all raw agent dialogues, proofs, and verification code, providing a transparent record of how these discoveries emerged.

Rel 85

August 26, 2026

How Warp builds self-improving agents on Claude

Warp addressed the limitations of stateless agent interactions by building a self-improving architecture on the Claude Platform. Initial attempts to refine prompts or context files such as AGENTS.md proved insufficient for scaling, as user feedback on outputs like code reviews typically vanished at the end of each session.

Rel 85

August 25, 2026

Google’s Spirit Data Deal Raises $10MN Question: Should AI Developers Be Allowed to Buy Enterprise Data for Training?

Google has bid $10MN to purchase data from bankrupt U.S. carrier Spirit Airlines, raising a broader question for enterprises. Should AI developers be able to buy the data businesses generate, particularly when that data contains years of operational information, employee activity and customer interactions? Court filings show that Google was selected as the successful bidder […]

Rel 85

August 25, 2026

SchemaRouter: Field-Aware Tool Routing for Efficient Heterogeneous Agentic RAG

Heterogeneous agentic retrieval-augmented generation (RAG) systems increasingly orchestrate external APIs, internal databases, vector stores, and graph stores. Exposing all tool descriptions to an LLM agent, or selecting tools only by vector similarity, causes two costly failures: over-fetching, which increases payload size, token use, and latency, and under-fetching, which omits fields needed to answer the query. We present SchemaRouter, a lightweight routing layer that represents tools, endpoints, parameters, response fields, domain concepts, units, provenance, and license policies as a schema graph. Given a query, SchemaRouter emits an executable tool plan specifying which tools to call and which fields to retrieve. A small LLM extracts intent, concepts, and source constraints, while field selection is deterministic over the graph through intent-group projection and concept-field matching with an alias layer. On a materials-science benchmark of 110 queries, SchemaRouter achieves answer accuracy of 0.71, matching fetch-everything within overlapping confidence intervals and exceeding prompt-all's 0.66, though their intervals overlap. It uses 227 retrieved-context tokens versus 2,066 for fetch-everything and achieves 2.7x lower end-to-end latency than prompt-all. It also obtains the best tool-exact rate of 0.93 and parameter validity of 1.0. SchemaRouter grounds provenance and license information in 62 percent of answers, compared with approximately 0 percent for all baselines. We also find that minimizing selected-field count is counterproductive: it reduces answer accuracy to 0.56 with negligible token savings, while recall-preserving projection restores top accuracy. SchemaRouter improves efficiency, schema-size-independent scaling, and verifiable provenance/license-grounded answering at competitive accuracy.

Rel 85

August 25, 2026

There Is No Neutral Harness: Modern LLM Leaderboards Are Manufactured by Config-Fragile Items

Multiple-choice benchmarks fix the questions and the correct answers, but not the harness: the order of the options, the wording of the prompt, and whether a language model's answer is read from generated text or from per-option likelihoods. Work on this harness sensitivity reports it as aggregate score variance, leaving unexamined which items the variance falls on and whether they are the items that separate one model from the next. We treat the evaluation harness of large language models (LLMs) as an independent variable and resolve its effect to single items. We introduce the \textit{fragility grid}: 12 open-weight instruction-tuned LLMs from 4 families answer the same 3{,}679 items from 4 benchmarks (ARC, HellaSwag, MMLU, TruthfulQA) under 26 equally defensible harness configurations, recording one correctness bit for every model, item, and configuration. The comparison is matched, since the items, the weights, and the greedy decoding stay fixed while only the harness varies. Under the grid a model's score is a band rather than a point: gemma4-31b scores between 31 and 89 percent depending only on the harness. Three results follow. On the items that two adjacent models both answer stably the pair is tied, and config-fragile items carry 95.7 percent of a pair's gap on average. Four of the 12 models reach rank one under some configuration, so the harness selects the winner. Item discrimination, the property that benchmark-compression methods maximize, correlates with fragility at 0.28 (95 percent CI 0.25 to 0.30), so compression keeps the fragile items rather than removing them. The scoring choice, not the option order that protocols usually fix, is the load-bearing axis. We release the per-item records and the analysis script, from which every number regenerates on a CPU in seconds, and we position the fragility grid as a check a leaderboard can run before it reports an order.

Rel 85

August 24, 2026

⚡ Weekly Recap: AI-Powered PLC Attacks, GitLab Attacks, Stripe Key Leaks and More

A package gets installed. A login prompt opens. A box sits exposed to the internet. Nothing looks unusual yet. That’s roughly the mood this week. Trusted tools turn hostile, old weak spots get fresh attention, AI makes exploit work cheaper, and researchers keep finding attacks that sound harder than they actually are. Plenty to clean up. Here’s the short version. ⚡ Threat of the Week U.S.

Rel 85

August 24, 2026

A Survey on Foundations and Frontiers of Multimodal Agentic Frameworks: Techniques and Applications

Advances in large language models (LLMs) have fueled a wave of research into agency: the ability to reason, plan, and act. This effort has produced agentic frameworks that orchestrate perception, memory, and decision-making around powerful LLM backbones. With the advent of large multimodal models (LMMs), these systems can process and integrate diverse modalities, including images, audio, and video, thereby improving their real-world applicability. Yet, while surveys of LLM-based agents exist, the role of multimodality in shaping agency has not been systematically examined in recent years. This survey fills the gap by analyzing the impact of multimodality across the core functional modules of the agentic framework: perception, reasoning, planning, memory, and action. Using this lens, we trace the evolution from text-centric agents to multimodal frameworks, examine how modalities are integrated through delegated, late-fusion, and early-fusion architectures, and assess the emergence of agentic behaviors enabled by grounded perception and multimodal reasoning. We organize existing work through a modality-centric taxonomy that links architectural design choices to agent capabilities. Moreover, we review multimodal agentic systems across various application domains, including Robotics, GUI & Web Navigation, Multimedia Content Generation & Editing, and Long-form Video Understanding & Retrieval. Beyond capabilities, we analyze performance across these settings and discuss efficiency-scalability trade-offs, including training and inference costs, latency, and deployment constraints. By focusing on the impact of multimodality in agentic design, we aim to identify key gaps and chart a roadmap toward robust and general-purpose intelligent systems.

Rel 85

August 24, 2026

When Retrieval Fails Before It Begins: Structurally Indirect Prerequisite Eviction as a Retention Failure in Agentic Memory

Agentic memory under a fixed budget involves two stages: retention and retrieval. Existing retrieval-centered paradigms implicitly assume necessary evidence survives eviction, but we challenge this by isolating a pre-retrieval failure mode: structurally indirect prerequisite eviction, in which upstream blocks weakly aligned with the query are discarded under budget pressure. We provide an operational definition of this failure, a reproducible deterministic benchmark, and per-seed trace diagnostics. Finally, we evaluate Dependency-aware Semantic Garbage Collection (DSGC), a one-hop graph-aware rule. In our main suite, DSGC improves full-chain retention from 0.03 to 0.90 under a lexical encoder and from 0.23 to 1.00 under a sentence encoder. Robustness checks then identify the budget and scaling regimes where the one-hop rule holds or degrades. Our released pipeline and failure postmortem support mechanistic analysis of retention before retrieval as a distinct failure boundary.

Rel 85

August 24, 2026

Nexus: Depth-Adaptive KV-Cache Splicing and Retrieval-Decoupled Tool Routing for Agentic LLMs on Unified Memory

Agentic large language models (LLMs) on the Model Context Protocol (MCP) re-encode verbose tool schemas every turn, so prefill - quadratic in sequence length - dominates time-to-first-token (TTFT) as the tool registry grows. Nexus's primary lever is to decouple routing from the schema-prefill cost: an INT8 semantic lookaside buffer (SLB) with a calibrated cross-encoder margin gate selects tools by retrieval, and arguments are generated over a compressed textual signature (median 19 tokens) rather than over spliced key/value (KV) cache. This path is depth-independent: routing accuracy stays near 89% as the registry scales to 250 tools - where a concatenate-all-schemas baseline overflows the context window entirely - and it reaches a first-argument token 1.66x sooner than a full-schema re-prefill at a ~80% main-context token saving. As a secondary, bounded lever we transplant a compiled schema KV block directly into the live context. This is fundamentally limited by rotary position embedding (RoPE) phase drift: an anchored splice is output-exact, but off-anchor placement corrupts attention, so beyond a threshold P=256 Nexus repairs the seam with a depth-adaptive suffix redecode that escalates to a full re-prefill. The resulting never-regress property is a guarantee on output fidelity (top-1 agreement, D_KL approx. 0) - not on latency, which can dip to 0.98x before converging to parity - alongside a 1.1-1.7x TTFT speedup at moderate depth that narrows to parity at deep context. Two negative results bound the design: the off-anchor RoPE fidelity boundary, and the failure of a reference-free drift gate to predict drift (Spearman rho = 0.193). All measurements are from one model tuple (Qwen2.5-14B-Instruct Q4_K_M) on Apple-silicon unified memory; the qualitative boundaries generalize, while the quantitative envelope is tuple-specific.

Rel 85

August 24, 2026

Interpretable Multimodal Classification with Linear Discriminant Tree Ensembles

Multimodal affect and behaviour classifiers that fuse heterogeneous text, audio, and visual streams must simultaneously achieve competitive accuracy and produce human-understandable explanations of the cues driving their decisions -- a dual objective that current high-capacity models, notably Transformers, only partially address. While Transformers attain strong predictive performance, their distributed representations and deep nonlinearity make it difficult to assign meaningful importance weights to individual multimodal features, limiting their use in trust-sensitive applications such as clinical affect monitoring and educational assessment. We address this gap by developing a framework based on tree-based ensembles that balances accuracy and interpretability. The framework encodes each modality into tokens, extracts and clusters concepts to reduce dimensionality, routes the fused modalities through tree-based ensemble classifiers, and interprets trends using a novel modified feature importance metric. The modified importance reduces the influence of the negative class in binary classification tasks, thereby improving indicator or marker detection. The proposed tree-based ensembles -- Linear Discriminant Tree (LDT), Linear Discriminant Forest (LDF), and Linear Discriminant AdaBoost (LDAB) -- achieve F1-mod gains of 4.3\% over the Multimodal Transformer and accuracy gains of 3.0\% over the primary interpretable multimodal baseline, Interpretable Multimodal Routing (IMR). The proposed multimodal feature importance extracts salient inter-modal concepts with substantially higher human-annotator agreement scores than default feature importance (62.2\% vs.\ 43.2\% on IEMOCAP; 46.7\% vs.\ 32.1\% on CMU-MOSI).

Rel 85

August 24, 2026

Truth Lies Deep: Countering Semantic Camouflage via Latent Intent Verification

Safety alignment in Large Language Models (LLMs) is often superficial, relying on refusal mechanisms that trigger only at the final stages of generation without erasing the foundational knowledge of harmful concepts acquired during pretraining. This study demonstrates that this architectural disconnect leaves models vulnerable to Semantic Camouflage -- adversarial attacks that wrap harmful intent in benign narrative contexts (e.g., creative writing), effectively bypassing standard input and output guardrails. By analyzing the latent activation trajectories of three distinct Small Language Model (SLM) families (Phi-3, Qwen2.5, and Gemma-2b) under adversarial stress, this research identifies a universal ``Intent Horizon'' -- a critical depth (typically 15--20\% of total layers) where the model's distinct, pre-trained representation of harmful intent collapses as it contextualizes the query into a ``safe'' narrative. Results indicate that while late-layer representations of camouflaged attacks are mathematically indistinguishable from safe queries (Detection Rate $< 20\%$), early-layer representations retain a distinct, detectable ``harm signature.'' Leveraging this insight, this paper proposes Latent Intent Verification (LIV), a lightweight probing defense. Experiments on the PKU-SafeRLHF dataset demonstrate that LIV outperforms standard guardrails by a margin of 20--50\% across all tested architectures, effectively neutralizing zero-day semantic attacks without requiring model retraining.

Rel 85

August 27, 2026

Amazon Kiro Prompt Injection Can Exfiltrate Sensitive Data Through Kiro Powers

Cybersecurity researchers have disclosed details of a vulnerability in Amazon Kiro, an artificial intelligence (AI)-powered, agentic integrated development environment (IDE), that could facilitate data exfiltration via prompt injection and Kiro Powers. The security flaw, which does not have a CVE identifier, works against Kiro IDE 0.7.45 on Windows, according to Mindguard. The latest version of

Rel 85

August 27, 2026

Google’s $10MN Spirit Deal Reveals AI Is Putting a Price on Enterprise and Customer Data

Google’s $10MN bid for the corporate data of bankrupt Spirit Airlines initially appeared to be an unusual consequence of a company collapse. The deal now looks more like the emergence of a developing market where AI companies are increasingly willing to pay large sums for the data generated by businesses and their employees. Google won […]

Rel 85