Topic
memory
Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents
A new research paper formulates memory retention in long-horizon language agents as a constrained stochastic optimization problem, proposing OSL-MR (Observability-Safe Learning for Memory Retention). The method combines an evidence learner with a Mixed-Score heuristic, achieving superior performance under tight budgets on benchmarks LoCoMo and LongMemEval. The work establishes a principled foundation for memory management in AI agents.
SelectStream: A Selective Memory Framework for Streaming Video Understanding
Researchers propose SelectStream, a selective latent-memory framework for streaming video models that uses surprise-driven adaptive windowing and query-conditioned graph reasoning to allocate memory efficiently. It achieves 82.67% on StreamingBench and 74.4% on offline benchmarks.
TS-Memory: A Plug-and-Play Memory Adapter for Time Series Foundation Models
TS-Memory, a lightweight memory adapter proposed by researchers, addresses distribution shift in time series foundation models. It uses Parametric Memory Distillation to combine the benefits of parametric adaptation and non-parametric retrieval without their drawbacks, achieving consistent forecasting improvements with minimal overhead.
Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection for Tool-Using LLM Agents
A new framework called the Counterfactual-Inspired Context Layer (CICL) helps LLM agents select and compress context based on decision relevance rather than semantic similarity. In tests on 50 SWE-bench Verified instances, CICL improved hit@1 from 0.58 to 0.78 and saved 44.93 tokens per query through memory cards.
User as Code: Executable Memory Paradigm Boosts AI Agent Personalization with Deterministic Execution
A new arXiv preprint introduces User as Code (UaC), a paradigm where AI agent user memory is stored as executable Python code rather than unstructured text. On benchmarks, UaC achieves 78.8% recall on LOCOMO and 99% accuracy on aggregate questions, while enabling unsolicited safety alerts that retrieval-based systems cannot provide.
E-mem: Multi-Agent Framework for Episodic Memory Reconstruction Boosts LLM Reasoning Efficiency by 70%
Researchers propose E-mem, a multi-agent framework that reconstructs episodic context for LLM agent memory, inspired by biological engrams. It uses a hierarchical architecture with assistant agents maintaining uncompressed contexts and a master agent orchestrating planning, achieving 54% F1 on the LoCoMo benchmark, surpassing the state-of-the-art GAM by 7.75% with over 70% token cost reduction.
Technology Old Nvidia GPUs Resurrected, SK Hynix Plans to Triple Memory Production by 2034
Old Nvidia RTX 3060 and 3050 GPUs are being resurrected by graphics card maker Manli in Asia as budget alternatives amid rising current-gen GPU prices. Meanwhile, SK Hynix chairman Chey Tae-won revealed plans to double wafer capacity by 2031 and triple it by around 2034, accelerating from a previous 20-year timeline to meet surging memory demand driven by AI.
Manufacturing Smartphone Brands Adjust Features Amid Rising Memory Costs
Smartphone manufacturers are adjusting features to cope with an 8-12% surge in memory costs. Brands like Xiaomi, Oppo, and Apple are compromising on display and audio quality to maintain price stability.