Mem2ActBench: A Benchmark for Evaluating Long-Term Memory Utilization in Task-Oriented Autonomous Agents

TL;DR

Mem2ActBench tackles the challenge of evaluating active long-term memory usage for grounding tool invocations in task-oriented agents. It constructs a memory-grounded benchmark by merging 2,029 long-context dialogues into coherent memory evolution chains and then reverse-generates 400 memory-dependent tool-use tasks, with human verification confirming 91.3% require memory. The authors evaluate seven memory frameworks across three Qwen2.5 backbones, finding that while retrieval matters, current systems struggle with grounding underspecified queries into exact tool parameters, especially for mid-history memories. The work highlights a critical gap between memory retrieval and memory-grounded action, and suggests research focus on improving evidence-hitting, memory organization, and grounding mechanisms to enable robust, long-horizon autonomous agent behavior.

Abstract

Large Language Model (LLM)-based agents are increasingly deployed for complex, tool-based tasks where long-term memory is critical to driving actions. Existing benchmarks, however, primarily test a angent's ability to passively retrieve isolated facts in response to explicit questions. They fail to evaluate the more crucial capability of actively applying memory to execute tasks. To address this gap, we introduce \textsc{Mem2ActBench}, a benchmark for evaluating whether agents can proactively leverage long-term memory to execute tool-based actions by selecting appropriate tools and grounding their parameters. The benchmark simulates persistent assistant usage, where users mention the same topic across long, interrupted interactions and expect previously established preferences and task states to be implicitly applied. We build the dataset with an automated pipeline that merges heterogeneous sources (ToolACE, BFCL, Oasst1), resolves conflicts via consistency modeling, and synthesizes 2,029 sessions with 12 user--assistant--tool turns on average. From these memory chains, a reverse-generation method produces 400 tool-use tasks, with human evaluation confirming 91.3\% are strongly memory-dependent. Experiments on seven memory frameworks show that current systems remain inadequate at actively utilizing memory for parameter grounding, highlighting the need for more effective approaches to evaluate and improve memory application in task execution.

Figures (5)

• Figure 1: Fact retrieval vs. memory-driven task execution. Existing benchmarks focus on direct queries for a factual answer. In contrast, our benchmark requires the agent to combine past memories and generate a grounded tool call.
• Figure 2: This diagram illustrates the Mem2ActBench framework, a benchmark used to evaluate the long-term memory capabilities of an agent. The framework first constructs a globally consistent and conflict-free "memory evolution chain" by integrating multi-source dialogue data. Then, based on this memory chain, it reverse-engineers question-answering tasks that require long-term memory to correctly select and use tools. Through this automated process, Mem2ActBench can effectively measure an agent's ability to proactively use its memory to complete tasks in complex, long dialogues.
• Figure 3: F1 score versus the normalized position of the earliest supporting memory.
• Figure 4: Breakdown of Slot Accuracy by Value Complexity (top) and Grounding Type (bottom).
• Figure 5: Distribution of failure modes across different memory frameworks.