Tool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a single prompt. The dominant approach, which we term Retrieve-then-Plan (RtP), embeds the user query and retrieves similar tools via approximate nearest-neighbor search. We study an alternative Plan-then-Retrieve (PtR) paradigm in which the LLM first generates Abstract Tool Stubs (hypothesized function signatures describing needed capabilities) which then serve as retrieval queries against the registry. Evaluating on ToolBench (46,984 unique API endpoints, 50 categories) across single-tool (G1), intra-category multi-tool (G2), and cross-category multi-tool (G3) query subsets, we find a clear complexity-dependent crossover: RtP outperforms PtR on single-tool queries while PtR outperforms RtP on multi-tool queries. This crossover holds at every retrieval budget (k=1 to 20), is robust across registry sizes from 1k to 47k APIs, and is consistent across model families (Gemini 2.5 Flash and gpt-oss-120b). We characterize a stub-hallucination failure mode (the LLM invents plausible but non-existent API names) that acts as a double-edged sword: on simple queries PtR’s stubs abstract away proper-noun signals that RtP matches directly, yet on complex queries the same mechanism decomposes multi-faceted intent into functional sub-needs. We further analyze token overhead and show that PtR’s extra discovery call roughly doubles inference cost, which pays off on multi-tool queries but hurts on single-tool ones, motivating complexity-adaptive routing as a practical direction — one that can be driven by a zero-cost signal already present in the pipeline (RtP’s top-1 retrieval score).

Characters and their interactions are central to the fabric of narratives, playing a crucial role in developing readers’ social cognition. In this paper, we introduce a novel annotation framework that distinguishes between five types of character interactions, including bilateral and unilateral classifications. Leveraging the crowd-sourcing framework of citizen science, we collect a large dataset of manual annotations (N=13,395). Using this data, we explore how genre and audience factors influence social network structures in a sample of contemporary books. Our findings demonstrate that fictional narratives tend to favor more embodied interactions and exhibit denser and less modular social networks. Our work not only enhances the understanding of narrative social networks but also showcases the potential of integrating citizen science with NLP methodologies for large-scale narrative analysis.

Non-Player Characters (NPCs) significantly enhance the player experience in many games. Historically, players’ interactions with NPCs have tended to be highly scripted, to be limited to natural language responses to be selected by the player, and to not involve dynamic change in game state. In this work, we demonstrate that use of a few example conversational prompts can power a conversational agent to generate both natural language and novel code. This approach can permit development of NPCs with which players can have grounded conversations that are free-form and less repetitive. We demonstrate our approach using OpenAI Codex (GPT-3 finetuned on GitHub), with Minecraft game development as our test bed. We show that with a few example prompts, a Codex-based agent can generate novel code, hold multi-turn conversations and answer questions about structured data. We evaluate this application using experienced gamers in a Minecraft realm and provide analysis of failure cases and suggest possible directions for solutions.