Partially observable reinforcement learning requires deciding what to retain, retrieve, and forget over time. We introduce a neuro-symbolic meta-policy that learns which symbolic memory heuristic to apply at each decision point while keeping execution symbolic. Our setting uses temporal knowledge-graph memory in RoomKG, where hidden state and observations are represented as Resource Description Framework (RDF) graphs and memory is augmented with temporal RDF triple annotations. The model combines knowledge-graph encoding of memory contents with value heads for question answering, exploration, and forgetting, yielding a controller that is both adaptive and inspectable. This gives the work a direct Semantic Web grounding through RDF-based representation, annotation-compatible graph semantics, and graph-based symbolic operations over explicit memory state. On train/test room splits at long-term memory capacity of 512, the qualifier-aware StarE-GNN configuration achieves the best held-out performance among the compared symbolic, neural, and neuro-symbolic systems while preserving step-level traceability of memory-management decisions.
Large language model (LLM) agents have demonstrated the ability to solve multi-step tasks requiring planning, tool use, and external information access, yet growing execution histories increase inference cost and expose reasoning to outdated, irrelevant, or misleading information, potentially degrading reasoning quality. Existing memory approaches organize or compress execution histories but provide limited mechanisms for deciding which memories remain active. We introduce the, a hierarchical memory system that organizes execution into tasks, subtasks, and actions while assigning each memory a dynamic retention score. Event-based updates and selection-based decay revise these scores, allowing WMT to preserve useful information, fold completed trajectories, suppress low-utility content, and retain access to folded context. We evaluate WMT on GAIA-Text using Qwen3-8B, Gemma 4 E4B, and Llama-3.1-8B, with ablations and memory-poisoning experiments. Relative to linear memory, WMT improves accuracy by an average of 9.97 percentage points while reducing prompt-token usage by 32.8%. Memory-poisoning experiments show that WMT limits the persistence and propagation of unreliable information. Our results suggest that effective long-horizon agent memory depends less on storing more information than on deciding which information should remain active.
Background: Transformer-based foundation models have achieved state-of-the-art results in various natural language inference benchmarks, but their decision-making processes remain largely unexplainable. Addressing the ’explainability gap’ is crucial for responsible AI adoption in highrisk industries that require transparency and trustworthiness. Furthermore, the combination of neural pattern matching with structured symbolic reasoning in resource-constrained scenarios is an important open problem. Objective: This study aims to present a memory-optimized probabilistic neuro-symbolic hybrid architecture that unifies transformer-based neural networks with logic-based symbolic reasoning systems. Methods: We use the e-SNLI dataset that provides human-written natural language explanations and reasoning highlights as training targets, and finetune the BERT transformer-based language model with an emphasis on gradient checkpointing, mixed-precision (FP16) training, and layer freezing for optimal resource utilization/reasoning tradeoffs. All experiments were performed on an NVIDIA GPU with 8–12 GB VRAM and CUDA-compatible hardware. Results: The proposed framework achieves 80.6% accuracy on 3-way NLI classification (contradiction, entailment, and neutral) with 0.806 precision, recall, and F1 scores on each class, and detailed class-level analysis shows high performance on entailment recognition (F1 = 0.912) and contradiction detection (F1 = 0.902), but slightly lower performance on neutral cases (F1 = 0.864). Ablation studies and confidence distributions of the model predictions indicate that memory-optimized models can maintain competitive performance and be deployed on resource-constrained devices, reducing GPU memory usage by ~60%. Conclusions: The results indicate that neuro-symbolic systems within memory-constrained systems can achieve both explanation needs and foundation models’ performance requirements, representing an important step in creating more trustworthy AI for NLP.
Zahraa Sameer Ibrahim, Haedar Ahmed Mukhef, Hayder Hasan Ali· Al-Mustansiriyah Journal of...· 0 citations
Recent advancements in inference-time scaling have significantly unlocked the complex reasoning capabilities of Large Language Models~(LLMs). However, for agents, these approaches suffer from a critical inefficiency, operating in a stateless manner and engaging in redundant search processes. Existing memory mechanisms largely rely on the reasoning capabilities of LLMs, leading to prohibitive computational costs. In this paper, we propose a novel framework, \textit{GAMER}~(Graph-based Action-centric Memory with Episodic Reasoning), that bridges the gap between inference scaling and episodic memory. Our approach models historical reasoning as a dynamic \textit{Action-Centric Graph}. By decoupling the memory mechanism from LLMs, our method can save token/money usage by providing less memory context than memory mechanism baselines. To extract knowledge from the graph effectively, we use a dual-stream Temporal Difference learning mechanism to estimate the positive~(suggestion) and negative~(avoidance) value of action nodes based on past successes and failures. During the inference phase, this learned value function optimizes decision-making bi-directionally, so that positive values provide action suggestions, while negative values indicate high-risk actions. By performing efficient searches on the graph, our method significantly improves the efficiency of inference scaling. Experiments on multiple benchmarks demonstrate that \textit{GAMER} achieves superior performance by \textbf{20.81\%/6.17\%} for success/progress rate compared to vanilla baselines.
Xu Zheng, Chaohao Lin, Zhuomin Chen et al.· 0 citations
Continuous-time dynamic graph models predict future links by compressing past interactions into neural states. Although effective for forecasting, this computation obscures which entities are shared across events and how temporal patterns contribute to a prediction. We treat this gap as a property of the predictive architecture rather than a problem to be addressed after prediction. Link-Fact Temporal Rule Inducer (LiFTER) is a neuro-symbolic predictor that preserves observed interactions as grounded temporal facts and applies executable tempo- ral rules to pre-query facts. Each score is a signed sum of rule exe- cutions whose historical facts, entity bindings, and temporal order are explicitly satisfied. The evidence and rules responsible for a prediction can therefore be inspected, independently recomputed, and intervened upon. Across four CTDG benchmarks, LiFTER achieves competitive historical-negative forecasting and the highest macro explanation ac- curacy and deletion fidelity. The same architecture also serves as a microscope that separates the contributions of recurrence, history po- sition, and transition across datasets and traces them to individual facts. Independent execution reconstructs all logits for 19,664 test predictions with a maximum error of 0.0000131. LiFTER turns future-link forecasting into a verifiable grounded computation.
Long-term memory is essential for language agents operating across extended interactions and evolving tasks. Existing memory-augmented agents mainly focus on storing and retrieving past experience, but the quality of stored memories may degrade over time. In particular, previously distilled insights can become outdated, over-generalized, or harmful under new task contexts, causing memory pollution when repeatedly reused. To address this issue, we study insight-level memory maintenance for long-term language agents and propose a failure-aware memory maintenance framework based on an editable insight graph. Each insight node tracks positive evidence, negative evidence, and an activation state, enabling the agent to distinguish reusable insights from conflicting or invalid ones. We further introduce a utility-aware retrieval mechanism and a graph controller that updates the memory graph after task execution by keeping reliable insights, archiving invalid ones, revising outdated ones, and adding newly discovered reusable insights. Extensive experiments show that our method consistently outperforms representative memory-based agent baselines across different backbone models. Ablation studies further demonstrate that append-only memory is insufficient for long-horizon tasks, while evidence-aware retrieval and graph-level editing improve memory reliability and downstream task performance.
As agentic tasks grow in complexity, LLM agents increasingly rely on experiential memory to reuse procedural knowledge across tasks. Effective memory design must jointly address what to store, how memory is structured and retrieved, and how memory evolves. Existing systems tackle each only partially: they store trajectories, insights, or workflows as isolated entries, discarding compositional relationships among subtasks and reusable skills; retrieve by flat embedding similarity that ignores relational signals; and maintain memory without leveraging its relational structure. We propose HyperSkill, a hypergraph-based memory framework that jointly improves all three. HyperSkill represents memory as a hypergraph with two node types, subtask steps and reusable skills, where each hyperedge links the subtasks and skills from a single trajectory. Dual-path retrieval queries both subtask and trajectory levels, ranking skills by co-occurrence across retrieved trajectories. Periodic structure-informed maintenance prunes low-utility nodes and merges redundant skills via quality-weighted propagation. Across xBench, GAIA, and WebWalkerQA with GPT-4o and Qwen3-30B-A3B, HyperSkill outperforms ten memory baselines, yielding gains of up to +11.51 on GAIA and +11.18 on WebWalkerQA.