Do Language Models Dream of Binding Molecules? Benchmarking LLMs under Spatial Constraints

  • 类型:arxiv
  • 标识:2607.18144
  • 链接:https://arxiv.org/abs/2607.18144
  • 主分类:evaluation
  • 形态:benchmark
  • 被引:0
  • 被引来源:Semantic Scholar + OpenAlex
  • S2被引:0
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:A clear pattern is revealed in LLM spatial capabilities: while they still lag behind state-of-the-art approaches, they are promising and can handle multiple spatial constraints simultaneously, enabling scaling to heterogeneous setups.
  • OpenAlex ID:W7169850691
  • OpenAlex DOI:10.48550/arxiv.2607.18144
  • DOI:10.48550/arxiv.2607.18144
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2607.18144
  • OpenAlex更新:2026-08-24
  • 待LLM分类:否
  • 标题中文:Do Language Models Dream of Binding Molecules? Benchmarking LLMs under Spatial Constraints
  • TLDR中文:研究揭示了 LLM 空间能力中的清晰规律:尽管其仍落后于 SOTA 方法,但具有潜力并能同时处理多种空间约束,从而可扩展到异构场景。
  • 来源文件
  • /inbox/tom/_candidates/2026-07-22-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]