收录解读
在科学、医学、法律、金融等复杂任务中,需要综合跨异构来源的海量信息,但现有模型上下文长度有限。现有方法通过图、文本记忆、检索集合等组织信息,但这些表示的设计和构建仍缺乏统一原则。
本文受认知科学中关联实现理论的启发,提出了设计大规模上下文表示的具体原则,并在此基础上开发了R3Con系统框架,用于系统化地实现这些原则。
在两项大规模文档推理基准上的实验表明,R3Con相比9个SOTA基线分别提升20%和8.4%的准确率,且允许小模型(4B/9B)超越更大模型(35B),甚至以更低成本超越闭源模型。
该工作为减少对模型规模的依赖提供了新路径,有望推动小模型在复杂推理场景中达到前沿水平。但实验仅覆盖两个基准,通用性待进一步验证。
原始摘要
Solving complex tasks in domains such as science, medicine, law, and finance often requires assembling interdependent information scattered across vast, heterogeneous sources far beyond model context limits. Existing approaches tackle this challenge by organizing information into more manageable representations over which models can reason, such as graphs, textual memories, and retrieval collections. These representations dictate what downstream reasoning is possible and, ultimately, whether it succeeds; yet their design and construction remain largely ad hoc. In this work, drawing on the cognitive theory of relevance realization, we propose concrete principles for designing AI systems that construct effective representations of very large contexts. We analyze existing approaches and show how their successes and failures map onto their alignment with these principles, and introduce R3Con, a harness designed to operationalize the principles more systematically. We evaluate R3Con against nine state-of-the-art baselines on two recent benchmarks of reasoning over large document corpora. On these benchmarks, R3Con substantially outperforms the strongest baseline, by $20$ and $8.4$ percentage points. It also enables smaller models to outperform much larger ones: R3Con with 4B and 9B models outperforms all evaluated 35B baselines, while R3Con with a 35B-A3B model outperforms Claude Code with Claude-Sonnet-5 at $3.7\times$ lower cost. Our results show that context representations following our principled approach can reduce reliance on model scale, pointing toward a future of AI systems with frontier-level performance powered by smaller models. Our code is available at https://github.com/michaeltheologitis/r3con