Paper page - The Laws of Context Allocation: Causal Measurement and Closed-Loop Orchestration in Generative Search

Source: https://huggingface.co/papers/2608.23252 We found that the metrics the RAG community uses to check “did the model actually use this document?” are mostly measuring the wrong thing.

On standard evaluation pools — where the negatives are passages retrieved for unrelatedqueries — BM25 and query–document cosine look essentially perfect at identifying the evidence a generator relied on (AUC 0.99). Swap the padding for same-query hard negatives that are topically dense but entail no answer, and they fall to chance (0.57). A causal leave-one-out probe barely moves (0.85 → 0.85). The apparent reliability was an artifact of the distractors being easy.

That matters because these proxies are what people use to decide what goes into the context. Once we measure utilization causally instead, a clear allocation law shows up: for a fixed budget of k×T evidence slots, spending it on several narrow sequential contexts beats one wide context by16.8–20.5 absolute pointsof portfolio recall, and the gap holds up to 32B. Widening the context mostly buys a slightly better single answer while leaving the rest of the answer space untouched.

We close the loop with a submodular scheduler that reads the probe’s feedback each round to demote saturated evidence and promote under-used evidence (+0.033 to +0.081 PR over every selection-style baseline, BH q < 0.001).

Code and the causal measurements are released — including an 11.5K-record benchmark where the necessary evidence set is known by construction, so you can check whether your own attribution method survives the swap from off-query to same-query negatives.

📄https://arxiv.org/abs/2608.23252 💻https://github.com/PeiYangLiu/ascp 🤗https://huggingface.co/datasets/PeiyangLiu/ascp-context-attribution