Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Entity trail

Big Bench Extra Hard

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

0

Findings

1

Edges

2

Sources

1

Corpus findings

2026-05-16 / arxiv-researcherPrompting Policies: RL-Trained Prompter Lifts Black-Box LLM Reasoning from 55% to 90%Google Research introduces a reinforcement learning framework that trains a lightweight prompter model to maximize task-specific rewards for a larger frozen worker LLM, using a contrastive experience buffer coupling scalar rewards with dense textual critiques. On Big Bench Extra Hard (BBEH) and Tau-bench, the approach improves logic-intensive reasoning from 55% to 90% and tool-use tasks from 74% to 91%. Key insight: prompt engineering can be amortized into learned single-shot policy weights rather than requiring iterative manual refinement.

Graph relationships

BENCHMARKED AGAINST
Prompting Policies
Prompting Policies improved reasoning on Big Bench Extra Hard from 55% to 90%.
Source finding
BENCHMARKED AGAINST
Prompting Policies -> Big Bench Extra Hard
Prompting Policies improved reasoning on Big Bench Extra Hard from 55% to 90%.
Source finding

Source trail

Graph sources

entity graphfindings textkg edgeskg entities

Big Bench Extra Hard intelligence trail | MindPattern