proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersWang Wei, Tiankai Yang, Samyadeep Basu1 min readpaperadvanced

Beyond Top-k Skill Retrieval: Diversity-Aware Skill Routing for LLM Agents

Read on arxiv.org

Summary

Beyond Top‑k Skill Retrieval: Diversity‑Aware Skill Routing (DSR) applies a Determinantal Point Process with a query‑residual diversity kernel to rerank skill candidates, balancing relevance and redundancy. On the SkillRouter benchmark it raises recall and full‑coverage, especially for multi‑skill queries, showing that skill routing benefits from set‑selection rather than independent ranking.

  • DSR uses a Determinantal Point Process to jointly optimize relevance and non‑redundancy among external skills.
  • A query‑residual diversity kernel penalizes overlap only when it harms complementary coverage, not merely shared relevance.
  • Benchmark results on SkillRouter show higher recall and full‑coverage than a strong pointwise reranker, with larger gains on multi‑skill queries.
  • The work reframes skill routing as a set‑selection problem rather than independent ranking of candidates.

LLM agents that call external tools need efficient skill selection to stay within context limits and avoid redundant calls, so a diversity‑aware router can improve performance on complex tasks.

6/10
llmml infraretrievalranking

Related reading

  1. The Router Within: Eliciting Native Skill Routing from a Frozen LLM

    The paper introduces Gavel, a method that extracts a frozen LLM's internal routing signal via two trained linear maps, eliminating the need to embed skill descriptions in the prompt. Experiments on Qwen3‑32B show up to 13.4‑point improvements on task benchmarks and higher skill‑use accuracy compared to larger retrieval‑based systems.

    Hugging Face Daily Papersarxiv.org1 minpaper

  2. CERA-MoA: Co-Evolving Routing Mechanisms with Continually Learning LLM Agents

    CERA-MoA proposes a reinforcement‑learning loop where a router and a set of LLM agents are trained together. A “familiarity” estimator reads mid‑layer hidden states to predict each agent’s competence on a query, letting the router activate only a minimal subset of agents that meet a cumulative confidence threshold. The system also feeds targeted training examples to agents based on their evolving…

    Hugging Face Daily Papersarxiv.org1 minpaper

  3. Online Learning with LLM Experts from Limited Feedback

    The paper models prompt routing to multiple LLM experts as a bandit problem with limited feedback and proposes algorithms that achieve sublinear regret in both full‑information and bandit settings. Experiments demonstrate that the methods learn effective routing strategies across diverse LLMs using only a small feedback budget.

    Hugging Face Daily Papersarxiv.org2 minpaper

  4. IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts

    IntBMoE introduces block‑level conditioning to MoE, decoupling token participation, compute execution, and memory materialization. A hypernetwork merges all experts into a composed expert per block, while routing remains sparse. Dual‑Path Residual Gating further mixes two composed paths. Experiments show consistent gains on vision, language, and recommendation tasks, and the model is live in AMap…

    Hugging Face Daily Papersarxiv.org2 minpaper

  5. COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization

    COBRA‑Skills uses a contextual‑bandit loop to selectively evaluate and evolve LLM agent skills, achieving better performance with roughly half the evaluation cost of prior methods. The framework works with limited examples and remains robust across different agent setups and even when the target model creates its own skills.

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Grounded Skill Synthesis from Code at Scale for Agentic Intelligence

    Code2Skill is an automated pipeline that mines popular GitHub repositories to extract verifiable, implementation‑anchored procedural “skills”. It builds a bank of ~1 M skill records (atomic ops, workflows, patterns) with provenance metadata, verifies each via blind reconstruction, and shows that augmenting LLM‑based agents with these skills yields an average 11.7% performance lift across 72 proto…

    Hugging Face Daily Papersarxiv.org1 minpaper