cd /news/artificial-intelligence/fcprag-fusion-controller-parametric-… · home topics artificial-intelligence article
[ARTICLE · art-109648] src=arxiv.org ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

FCPRAG: Fusion-Controller Parametric Retrieval-Augmented Generation for Stable Multi-Passage LoRA Injection

Researchers propose FCPRAG, a fusion-controlled parametric retrieval-augmented generation framework that adds a lightweight controller for sample-level adapter fusion, improving multi-passage LoRA injection stability. On HotpotQA, 2WikiMultiHopQA, PopQA, and ComplexWebQuestions across three LLM backbones, FCPRAG improves F1 by up to 4.65% on 2WikiMultiHopQA and 7.55% on CWQ over standard RAG and parametric RAG baselines, while reducing tuning cost and improving robustness.

read1 min views1 publishedAug 25, 2026

arXiv:2608.21750v1 Announce Type: new Abstract: Parametric retrieval-augmented generation (PRAG) injects retrieved evidence into a large language model (LLM) through passage-specific LoRA adapters, reducing reliance on long in-context prompts. When multiple passages are retrieved for the same query, however, evidence-level fusion becomes a bottleneck: equal-weight merging can amplify weak or conflicting evidence, and translating retrieval signals into fusion weights often requires fragile global tuning. We propose FCPRAG, a fusion-controlled parametric RAG framework that adds a lightweight controller for retrieval-conditioned, sample-level adapter fusion. The controller predicts per-passage fusion scores together with sample-level calibration signals, including a mixing gate and an adaptive temperature, enabling fusion that stays selective under informative retrieval signals and conservative under uncertainty. FCPRAG is trained with merge-aware supervision derived from each adapter's marginal contribution within a multi-adapter merge, using training data only. We further show that a single dataset-level temperature is suboptimal under heteroscedastic retrieval uncertainty, motivating sample-level adaptation. Experiments on HotpotQA, 2WikiMultiHopQA, PopQA, and ComplexWebQuestions (CWQ) across three LLM backbones show that FCPRAG consistently improves F1 over standard RAG and parametric RAG baselines, with gains of up to 4.65% on 2WikiMultiHopQA and 7.55% on CWQ, while also reducing tuning cost and improving robustness under retrieval perturbations.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @fcprag 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/fcprag-fusion-contro…] indexed:0 read:1min 2026-08-25 ·