Weekly Recap

This work introduces ACLArena, a framework for comprehensively studying, analyzing, and evaluating Agent Continual Learning, and proposes a new ACL recipe that combines offline replay over high-quality trajectories with a routed network of multiple LoRA experts each specialized via RL, substantially improving the agent's ability to learn across multiple domains.

This work introduces ACLArena, a framework for comprehensively studying, analyzing, and evaluating Agent Continual Learning, and proposes a new ACL recipe that combines offline replay over high-quality trajectories with a routed network of multiple LoRA experts each specialized via RL, substantially improving the agent's ability to learn across multiple domains.

A teacher-student framework combining an improved visual encoder architecture, persistent feature masking, and a novel consensus-gating algorithm to retain useful depth information under unmodeled corruption, which improves robustness where augmentation coverage is missing.

This work presents PARTE (Plane-Assisted Robust Transformation Estimation), a global registration method that instead treats planar structure as complementary registration evidence and achieves the highest overall success rate against 13 standard and state-of-the-art methods while maintaining low runtime.

A conditional decision-preservation proposition: successful local admission implies that a specified synchronous policy would authorize the same action at the admission point, provided approval is sound, all policy dependencies are represented and current, observations are faithful, and consumption is atomic.

This work recast black-box black-box LLM steganography as a sequential communication problem with causal, noiseless feedback: every generated token is observed by both parties and can guide subsequent embedding.

Several open research questions are identified, including how to efficiently record incidents and how to determine whether vulnerabilities and incidents generalize, and privacy requirements are summarized and research directions for the secure and trustworthy deployment of AI agents are outlined.

It is concluded that cross-paper ASR comparison is currently unsupported, and a ten-item reporting checklist targeted at each failure is proposed, to supply the shared measurement contract the field has so far done without.

A training-free agentic framework built around specialist detectors, per-detector triage, and conflict-aware evidence arbitration is studied, indicating that the main challenge of open-world image forensics is not detecting manipulations, but calibrating trust in specialized forensic tools and arbitrating conflicting evidence.

LeaseGuard is presented, a deterministic admission layer that represents preemption authority through canonical resource leases, incumbent-health checks, effect-aware admission, effect-aware admission, coexistence limits, safe alternatives, and resource-scoped overrides before adapter execution.

A framework for evaluating LLM-generated RTL beyond compilation, which is named GRADE-RTL, provides a practical basis for comparing LLM-generated hardware descriptions by separating structural validity, behavioral correctness, and implementation quality.

It is shown that a small, reusable set of optimization patterns---loop flattening, $128$-bit vectorized loads, splitting fused kernels, and profile-guided launch configuration---recovers CUDA-class performance without changing the public API, and argued that these patterns constitute a practical recipe for performance-portable, vendor-agnostic HEP analysis kernels.

This paper proposes SVEET, a framework that requires merely training on a pretrained bidirectional video diffusion model but supports high-quality streaming video editing in an auto-regressive fashion and proposes a decoupled training scheme that explicitly enforces the orthogonality between the optimization directions of video controllability and model causality.

This work adds one feature-alignment term to ordinary VLA training: a frozen world model is run over the training frames once and cached, and the student learns to agree with that cache, indicating a broad representational prior rather than a fragile alignment between two particular networks.

Unalloc, an open-source tool that joins OpenCost, LiteLLM, OpenAI and Anthropic cost data into one exact ledger and reports the share of spend with no owner, is presented to study where attribution breaks at the seams between these systems.

This work investigates GPU-resident lossy compression, using representative fields from the DYAMOND high-resolution operational dataset as input, and combining measured GPU compression throughput with SimGrid network simulation, and shows that ZFP at 16 bits per value (rate-16), about the same storage budget as float16, matches the communication-time reduction of float16 truncation.

ScaleMPA, a motion-planning accelerator that rethinks RRT* with a grid-native representation, achieves millisecond-level planning latency and delivers 4.4 times speedup over state-of-the-art motion-planning accelerators.

Qwen-Audio-3.1-Realtime brings these requirements together through Think, Act, and Speak and Coordinate, evaluating audio reasoning, multilingual understanding, tool use, conversational behavior, full-duplex interaction, and safety.

ReFilter is a hybrid framework that first Retrieves semantically related candidate apps using embeddings and then applies LLM-based contextual Filtering to identify true functionally similar apps with higher precision, enabling more accurate app comparisons and supports improved ecosystem understanding, competitor analysis, and recommendations.

It is recommended that frontier AI developers begin hosting embedded assessments now, covering at least three areas central to managing risks from internal AI use: internal agent monitoring, internal agent security controls and permissions, and model alignment.

A conceptual framework for participatory speech data curation by, for, and with marginalized communities drawing on insights from co-design and knowledge sharing is developed, comprising overlapping and two-way processes of defining a community, project formulation, modes of participation, and personal autonomy.

The findings to date suggest that significant abrupt changes in climate-change stance do occur in this simple model, and the need to prevent LLM biases from interfering with the conversational dynamics and, more generally, to maintain agent personality and episodic memories of interactions in the face of such biases is reported.

Convex AI compositionality is introduced: a formal representation of the configurations generated by finite AI system populations that uses convex spaces that are compatible with lifecycle reachability across the population and can preserve the formal identity relations between these systems.

This work shows that indirect tipping through intermediate stepping-stone equilibria can reduce the committed minority required to reach an alternative state, bypass majority requirements, and make possible transitions inaccessible through direct challenges.

A network-based framework that estimates influence directly from retweet networks using the independent cascade model and a polynomial-time dynamic programming algorithm that computes the optimal placement of coordinated nodes, providing an upper bound on their achievable influence are introduced.

A comprehensive taxonomy is proposed that maps physical, network, and application architectural IoT layers against the consumer, commercial, industrial, and infrastructure IoT domains and identifies key gaps in current research and outline future directions for context-aware and adaptive trust management in edge-enabled IoT.

Document Retrieval-Aware Chunking (D-RAC), an extension of the Web Retrieval-Aware Chunking (W-RAC) framework to arbitrary document formats, is presented, preserving W-RAC's cost, determinism, and observability benefits while unlocking every renderable format as a first-class input.

This work proposes CARE (Corrective Atomic Robotic Execution), a framework that improves recovery by learning from failures encountered during execution, and introduces the Failure State Recovery Benchmark (FSR-Bench), which evaluates recovery from intermediate failure states under local deviations and structural anomalies.

OnPanda is presented, an interactive tool for efficiently annotating LLM alignment data and agent trajectories that adopts token-level correction as its core interaction and releases Panda-CVL, a dataset annotated with onPanda, together with a benchmark for token-level correction.

This work introduces Harness-Zero, which enables harness distillation through agent-as-harness, and shows that for frontier LLMs using the same evolved harness, agent-as-harness outperforms code-as-harness.

VideoGen-Agent is presented, a multimodal agent trained through multitask agentic reinforcement learning to use external tools for video generation to support learning tool use across video-generation tasks and show that the trained agent can benefit from subsequent advances in generation tools.

The GameHorizon Suite, a unified data and evaluation suite that measures gameplay capabilities at different horizons for diverse model families, can provide a standardized yardstick for evaluating gameplay capabilities across horizons and model families.

Regularized Recursive Self-Improvement of Agent Harnesses (RRSI) is introduced, which incorporates the principles of regularizations into harness self-improvement by constraining the evolution candidate proposal and selection.

This work shows that Kimi Delta Attention (KDA) can realize 2D rotations by combining a single delta-rule transformation with a second reflection supplied by its channel-wise gate, and characterize the expressivity of CKDA and prove that every orthogonal diagonal-plus-rank-one matrix is exactly a CKDA transition matrix.