In on-device agentic systems, Small Language Models (SLMs) are essential for ensuring privacy-preserving personalization and low-latency execution. However, despite their capacity to handle isolated atomic tasks, during complex multi-step orchestration SLMs frequently suffer from stochastic drift—a phenomenon where minor logical deviations cascade into irrecoverable trajectory failures. In this work, we demonstrate the primacy of pre-execution planning, revealing that mitigating initial logical drift is vastly more effective than mid-trajectory interventions. Building on this insight, we propose FORST(FOundation fiRST) framework that implements sequential logical scaffolding in pre-execution planning, where predefined expert anchors are injected as fixed pre-computed KV-cache states. While the scaffolding ensures logical stability, KV-cache reuse efficiently realizes the structured anchor prefix without repeatedly prefilling static anchors. Then, we can maintain a monolithic reasoning chain within predefined logical planning boundaries and effectively eliminate the redundant computational overhead of structured anchors. Extensive evaluations on the MCPEval benchmark demonstrate that establishing this rigid logical foundation via achieves competitive or numerically higher performance than frontier models, while achieving an latency reduction on edge-device platforms.

Personal memory agents process sensitive user data, making on-device deployment with small language models (SLMs) attractive. However, existing memory-agent prompts are often monolithic specifications written for high-capacity LLMs, overloading a single inference with many requirements while leaving necessary intermediate operations implicit. This mismatch creates two bottlenecks for SLMs: prompt complexity and logical incompleteness. We introduce CAST (Compiling Agent Specifications into Scaffolded Tasks), a framework that converts a monolithic memory-agent prompt into a scaffolded execution procedure. CAST segments the original prompt into fixed, semantics-preserving main tasks, then synthesizes and places auxiliary scaffolding tasks that produce reusable intermediate artifacts; a trajectory-feedback step then adapts these scaffolds to the target SLM. At inference time, only the target SLM executes the compiled procedure. Across two memory-agent frameworks on LoCoMo and MemoryAgentBench, CAST outperforms monolithic prompting, prompt optimization, and reasoning-based baselines, while offering a stronger quality-latency trade-off than unconstrained reasoning. The results show that adapting memory agents to SLMs requires restructuring execution, not merely rewriting prompts.

Retrieval-Augmented Generation (RAG) systems are typically evaluated under static assumptions, despite being frequently corrected through user or expert feedback in deployment. Existing evaluation protocols focus on overall accuracy and fail to capture how systems adapt after feedback is introduced. We introduce feedback adaptation as a problem setting for RAG systems, which asks how effectively and how quickly corrective feedback propagates to future queries. To make this behavior measurable, we propose two evaluation axes: correction lag, which captures the delay between feedback provision and behavioral change, and post-feedback performance, which measures reliability on semantically related queries after feedback. Using these metrics, we show that training-based approaches exhibit a trade-off between delayed correction and reliable adaptation. We further propose PatchRAG, a minimal inference-time instantiation that incorporates feedback without retraining, demonstrating immediate correction and strong post-feedback generalization under the proposed evaluation. Our results highlight feedback adaptation as a previously overlooked dimension of RAG system behavior in interactive settings.

Large Vision-Language Models (LVLMs) are promising scalable annotators for video understanding, but they are not directly suited for dense temporal annotation. We study this challenge in egocentric streaming understanding, where the goal is to segment continuous first-person video into atomic actions: short, concrete physical interactions with well-localized temporal boundaries.

We propose CoTT (Chains of Temporal Transitions), an annotation-free framework that structures LVLM reasoning around temporal transitions and combines it with iterative streaming distillation. CoTT uses structured LVLM reasoning to generate dense atomic action pseudo-labels from unlabeled egocentric videos, distills these labels into a lightweight online model, and bootstraps the model’s predictions to improve temporal consistency.

On large-scale egocentric benchmarks, EPIC-Kitchens-100 and Ego-Exo4D, CoTT achieves competitive performance against supervised online localization methods without using ground truth temporal annotations for training. It also transfers across egocentric datasets and, on THUMOS14, outperforms prior best training-free zero-shot Temporal Action Localization methods, even without any dataset-specific training or adaptation.