AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Thinking Forward and Backward: Multi-Objective Reinforcement Learning for Retrieval-Augmented Reasoning

Wenda Wei, Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Lixin Su, Shuaiqiang Wang, Dawei Yin, Maarten de Rijke, Xueqi Cheng

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add backward reward and interpolate with forward model (ablation)Forward-RAR (Qwen2.5-3B-Base) · 7 QA benchmarks Avg.Exact Match: 0.333 -> 0.338Improved
add backward reward and interpolate with forward model (ablation)Forward-RAR (Qwen2.5-3B-Instruct) · 7 QA benchmarks Avg.Exact Match: 0.384 -> 0.397Improved
add forward reward and interpolate with backward model (ablation)Backward-RAR (Qwen2.5-3B-Base) · 7 QA benchmarks Avg.Exact Match: 0.321 -> 0.338Improved
add forward reward and interpolate with backward model (ablation)Backward-RAR (Qwen2.5-3B-Instruct) · 7 QA benchmarks Avg.Exact Match: 0.377 -> 0.397Improved
use forward (answer-side) reward instead of backward reward (ablation)Single-reward RAR (Base) · 7 QA benchmarks Avg.Exact Match: 0.321 -> 0.333Improved
use forward reward instead of backward reward (ablation)Single-reward RAR (Instruct) · 7 QA benchmarks Avg.Exact Match: 0.377 -> 0.384Improved
replace outcome-only reward with bidirectional information-distance rewardsSearch-R1 (Base) · 7 QA benchmarks Avg.Exact Match: 0.312 -> 0.338Improved
replace outcome-only reward with bidirectional information-distance rewardsSearch-R1 (Instruct) · 7 QA benchmarks Avg.Exact Match: 0.336 -> 0.397Improved
replace outcome-only reward with bidirectional rewardsSearch-R1 (Instruct) · HotpotQAExact Match: 0.331 -> 0.391Improved
replace outcome-only reward with bidirectional rewardsSearch-R1 (Instruct) · 2WikiMultiHopQAExact Match: 0.310 -> 0.402Improved
replace outcome-only reward with bidirectional rewardsSearch-R1 (Base) · BamboogleExact Match: 0.128 -> 0.188Improved
replace outcome-only reward with bidirectional rewardsSearch-R1 (Instruct) · BamboogleExact Match: 0.232 -> 0.363Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy