SPARK is a corrective multi‑episode framework for embodied retrieval‑augmented generation that uses a semantic critic, a targeted planner, and selective merging to refine a knowledge base after each episode. In experiments on four AirSim outdoor maps under degraded viewpoints, a single corrective episode improves embodied question‑answering performance on every map while reducing stored objects by 13–41 %. Extending correction to four episodes surpasses the performance of the undegraded reference on all maps, and ablations confirm the importance of each component.
arXiv:2609.35808v1 Announce Type: cross Abstract: Experience reuse can reduce repeated exploration in embodied agents, but a trajectory that succeeded previously may be unsuitable for the current exe...
MemoProAgent tackles memory‑driven proactive assistance by mining long‑horizon personal memories to decide when an assistant should intervene. The authors introduce MemoProBench, a benchmark pairing egocentric video with year‑long, distractor‑rich user memories and human‑verified trigger windows, and show that existing models perform poorly (≤15%). MemoProAgent uses a graph‑organized memory store and a lightweight RL stage to improve accuracy from <10% to 42.4% without training.
The paper introduces MERIT, a benchmark that evaluates the marginal benefit of long‑term memory for tool‑using large language model agents while explicitly accounting for cost. MERIT provides episodic tool‑use tasks across three domains, verifies dependence on earlier‑episode facts, and measures memory operations in tokens and dollars. Experiments on GPT‑4.1‑mini, Claude Haiku 4.5, and Claude Sonnet 5 show that memory can significantly improve task success, but its utility varies widely across models and memory implementations, and full replay is rarely cost‑effective.
arXiv:2610.23372v1 Announce Type: cross Abstract: When we say that a robot uses memory, does that memory actually play a role? Success-rate ablations alone cannot establish the use of task-relevant f...
The paper introduces ProPerMem, a benchmark for memory-driven proactive assistance that pairs egocentric video with long, distractor-rich personal memories and human-verified trigger windows. Baseline models perform poorly, achieving at most 15% accuracy, while an oracle-driven GPT-5.3-Chat model reaches 96% when given evidence inside the trigger window. Building on these insights, the authors develop ProPerAgent, a plug‑and‑play system that retrieves relevant memory from a graph store and uses lightweight reinforcement learning to improve proactive intervention accuracy to 42.4% without training.
arXiv:2610.08586v1 Announce Type: new Abstract: Long conversational agents have become essential in our daily lives. They must remember what was said long back in order to help us efficiently complet...
The paper introduces Multimodal Filesystem Memory (MFM), a memory architecture for long‑horizon embodied agents that stores episodic, spatial, semantic, procedural, and working memories. MFM uses searchable textual records to link these representations to multimodal episodes, enabling agents to retrieve past experiences via contextual cues and revisit original observations for missing details. Experiments on long‑horizon Minecraft and computer‑use tasks show that MFM improves performance over three base models, achieving higher milestone completion and reward scores compared to baselines.
MemoryVLN introduces a memory-driven navigation system that compresses long navigation histories and reuses past spatial experience. It uses forgetting-curve-inspired temporal sampling for long sequence compression, an ego-aware memory representation that fuses 2D semantic, 3D geometric, and 6D pose tokens, and a hierarchical memory bank for in-episode and context-level memory management. The system retrieves top‑K relevant memories to augment current visual tokens, achieving significant improvements on R2R‑CE and RxR‑CE benchmarks.
As assistants move from reactive answering toward proactive intervention, memory changes role: it no longer only supplies the answer to a query, it determines whether and when to speak at all. Prior w...
arXiv:2606. 29914v1 Announce Type: cross Abstract: Agent memory systems are increasingly evaluated against RAG and full-context baselines, but reported gains often mix changes in the memory method with changes in the language model, embedding model, or retrieval pipeline, making it unclear what is actually being measured.
arXiv:2609.28236v1 Announce Type: new Abstract: Long-horizon embodied interaction requires agents to retain and continually update information about the environment as they observe, act, and encounte...
arXiv:2610.33487v1 Announce Type: cross Abstract: Robots can use historical observations to answer spatiotemporal queries in familiar environments, but the presence of relevant evidence does not impl...