GaiaFlow: Semantic-Guided Diffusion Tuning for Carbon-Frugal Search

Rong Fu; Jia Yee Tan; Chunlei Meng; Shuo Yin; Xiaowen Ma; Wangyu Wu; Simon Fong

cs.IRFeb 17, 2026

GaiaFlow: Semantic-Guided Diffusion Tuning for Carbon-Frugal Search

Authors: Rong Fu, Jia Yee Tan, Chunlei Meng, Shuo Yin, Xiaowen Ma, Wangyu Wu, Simon Fong

Organizations: Independent Researcher

Abstract

As the burgeoning power requirements of sophisticated neural architectures escalate, the information retrieval community has recognized ecological sustainability as a pivotal priority that necessitates a fundamental paradigm shift in model design. While contemporary neural rankers have attained unprecedented accuracy, the substantial environmental externalities associated with their computational intensity often remain overlooked in large-scale deployments. We present GaiaFlow, an innovative framework engineered to facilitate carbon-frugal search by operationalizing semantic-guided diffusion tuning. Our methodology orchestrates the convergence of retrieval-guided Langevin dynamics and a hardware-independent performance modeling strategy to optimize the trade-off between search precision and environmental preservation. By incorporating adaptive early exit protocols and precision-aware quantized inference, the proposed architecture significantly mitigates operational carbon footprints while maintaining robust retrieval quality across heterogeneous computing infrastructures. Extensive experimental evaluations demonstrate that GaiaFlow achieves a superior equilibrium between effectiveness and energy efficiency, offering a scalable and sustainable pathway for next-generation neural search systems.

Figures & tables

Figure 1: Overview of the GaiaFlow framework for semantic-guided, carbon-frugal search optimization. An incoming query $q$ is encoded by the Input & Retrieval Tower into a semantic embedding $u_{q}$ . In the Latent Manifold , a Retrieval-Guided Langevin Sampler explores the configuration space $\mathcal{Z}$ , driven jointly by the Carbon-Frugal Gradient $\nabla_{z}U$ from the Multi-Objective Engine (Diff-PEIR) and the Semantic Attraction $\nabla_{z}V$ that maintains high-quality retrieval behavior. Real-time efficiency is supported by Quantized Inference and an Early Exit mechanism. The Online Calibration Loop uses EW-RLS and PUE correction to adapt to datacenter conditions, while the Operational Safeguard validates, projects, and finalizes the deployed configuration $\omega^{\ast}$ .

Figure 2: Measured versus predicted carbon emissions. Each point corresponds to a query instance. The dashed diagonal denotes perfect prediction, while the shaded region indicates a $\pm 10\%$ relative error band. Results show that the predictor remains well-calibrated across a wide dynamic range.

Figure 3: Latency and computational cost (Mop) distributions across retrieval methods. Boxplots summarize per-query latency (left axis) and Mop cost (right axis). GaiaFlow consistently achieves lower latency and computational overhead compared to baseline methods.

Figure 4: Coefficient distributions under full-query training and a $5\%$ subsample. Solid curves correspond to the full dataset, while dashed curves denote the subsampled setting. The strong alignment across coefficients indicates robust distributional stability under aggressive subsampling.

Figure 5: Pareto frontier over system configurations. Each point represents a deployment configuration, plotting average carbon cost against average latency, with color encoding retrieval recall. The frontier highlights favorable trade-offs achievable by GaiaFlow .

Figure 6: Online calibration behavior over time. The curve reports the mean absolute error (MAE) of latency prediction as new observations are incorporated. The decreasing trend indicates effective self-calibration during continuous deployment.

Figure 7: Representative Langevin trajectory during configuration optimization. The potential energy $U(z)$ decreases rapidly in early iterations, illustrating fast convergence and enabling early stopping in practice.

Retrieval method	Intel					AMD
	Mop	Flop	MFlop	LR	PL	Mop	Flop	MFlop	LR	PL
BM25 [ 61 ]	0.801	-0.259	0.969	0.990	0.589	0.463	-1.076	0.820	0.884	0.537
BM25-T5 [ 62 ]	0.968	0.367	0.989	0.997	0.000	0.742	0.308	0.950	0.964	0.001
uniCOIL-TILDE [ 63 ]	0.986	0.930	0.985	0.984	0.300	0.979	0.930	0.977	0.977	0.302
DeepImpact [ 64 ]	0.969	0.469	0.987	0.993	0.444	0.957	0.485	0.970	0.977	0.429
uniCOIL-T5 [ 65 ]	0.971	0.939	0.985	0.977	0.261	0.966	0.941	0.981	0.973	0.255

Table 1: R 2 scores for different performance models and retrieval methods.

Method	L_AMD	L_Intel	Mop [ $\times 10^{6}$ ]	L_AMD / BM25	L_Intel / BM25	Mop / BM25
BM25 [ 61 ]	13.90	12.26	5.4116	1.00	1.00	1.00
BM25-T5 [ 62 ]	17.15	18.32	7.3242	1.23	1.49	1.35
DeepImpact [ 64 ]	24.50	26.37	17.9721	2.49	2.60	2.70
SPLADEV2 [ 66 ]	277.97	290.97	161.8180	28.24	28.66	24.28
uniCOIL-T5 [ 65 ]	48.92	50.69	32.4773	4.97	4.99	4.87
uniCOIL-TILDE [ 63 ]	40.88	41.32	26.5466	4.15	4.07	3.98

Table 2: Comparison of retrieval methods: average latency (ms) and average Mop.

Configuration	Latency (ms)	Mop $\!\times\!10^{6}$	Recall@1000	Carbon $\!\times\!10^{-3}$ g	Steps	$\Delta$ Recall
GaiaFlow	9.0	4.00	0.859	2.18	6.1	0
GaiaFlow ${}_{\gamma_{2}=0}$	9.9(+10%)	4.38(+9.5%)	0.857(-0.2%)	2.39(+9.6%)	7.0(+15%)	$-$ 0.2%
GaiaFlow ${}_{\gamma_{1}=0}$	11.3(+26%)	4.95(+24%)	0.860(+0.1%)	2.70(+24%)	9.5(+56%)	$+$ 0.1%

Table 3: Impact of removing retrieval-guided attraction on MS-MARCO dev. Changes are measured relative to the complete GaiaFlow model.

Setting	Latency (ms)	Mop $\!\times\!10^{6}$	Recall@1000	$\Delta$ Carbon
$\gamma_{1}=1.8\times 10^{-3}$	9.6 $\pm$ 0.3	4.27 $\pm$ 0.05	0.858 $\pm$ 0.001	+6.5%
$\gamma_{1}=\mathbf{2.0\times 10^{-3}}$	9.0 $\pm$ 0.2	4.00 $\pm$ 0.03	0.859 $\pm$ 0.001	—
$\gamma_{1}=2.2\times 10^{-3}$	8.7 $\pm$ 0.2	3.89 $\pm$ 0.04	0.859 $\pm$ 0.001	$-$ 3.2%
$\gamma_{2}=1.8\times 10^{-3}$	9.3 $\pm$ 0.2	4.11 $\pm$ 0.04	0.858 $\pm$ 0.001	+2.4%
$\gamma_{2}=\mathbf{2.0\times 10^{-3}}$	9.0 $\pm$ 0.2	4.00 $\pm$ 0.03	0.859 $\pm$ 0.001	—
$\gamma_{2}=2.2\times 10^{-3}$	9.1 $\pm$ 0.2	4.03 $\pm$ 0.04	0.859 $\pm$ 0.001	+0.8%

Table 4: Sensitivity of individual step coefficients on MS-MARCO dev. Default values are shown in bold.

Train $\rightarrow$ Test	$R^{2}$	MAE (ms)
AMD $\rightarrow$ AMD	0.995	0.38
AMD $\rightarrow$ Intel	0.987	0.42
Intel $\rightarrow$ Intel	0.993	0.40
Intel $\rightarrow$ AMD	0.986	0.44

Table 5: Portability of the learned latency model across CPU architectures. “Train

\rightarrow

Test” denotes the source and target processors.

Information Retrieval

Information Retrieval studies how to find, rank, and organize relevant information in large collections of data.

See all 471 papers

GAR: Carbon-Aware Routing for LLM Inference via Constrained Optimization
May 12, 2026Disha Sheshanarayana, Rajat Subhra Pal, Manjira Sinha +1LLM Inference Efficiency Constrained Optimization
Manipal University Jaipur · TCS Research
Designing Sustainable Federated Learning as a Service using Neural Architecture Search
Aug 14, 2026Keya Patel, Sajib Mistry, Sheik Fattah +4Federated Learning Neural Architecture Search
School of Electrical Engineering, Computing and Mathematical Sciences Curtin University, Perth, Australia · King Fahd University of Petroleum and Minerals, Saudi Arabia · Umeå University, Umeå, SE-90187, Sweden
EcoThink: A Green Adaptive Inference Framework for Sustainable and Accessible Agents
Mar 26, 2026Linxiao Li, Zhixiang LuLLM Inference Efficiency Energy-Efficient ML
The University of Sydney Sydney, Australia · University of Liverpool Liverpool, United Kingdom

May 12, 2026J/K move · Enter open · S save

GAR: Carbon-Aware Routing for LLM Inference via Constrained Optimization

Disha Sheshanarayana, Rajat Subhra Pal, Manjira Sinha, Tirthankar Dasgupta

Manipal University Jaipur · TCS Research

LLM Inference Efficiency Constrained Optimization LLM Routing Carbon-Aware Computing

Open paper AskPDF arXiv

The growing deployment of large language models (LLMs) makes per-request routing essential for balancing response quality and computational cost across heterogeneous model pools. Current routing methods rarely consider sustainable energy use and CO2 emissions as optimization objectives, despite grid carbon intensity varying by time and region, and models differing significantly in energy consumption. To address this gap, we introduce Green-Aware Routing (GAR), a constrained multi-objective optimization framework that minimizes per-request CO2 emissions subject to explicit accuracy floors and p95-latency service-level objectives (SLOs). GAR employs adaptive constraint optimization through per-dataset floor tuning and incorporates lightweight estimators for correctness, tail latency, and carbon emissions, enabling real-time routing decisions without additional inference passes. We present GAR-PD, a practical online primal-dual routing algorithm for rolling carbon budgets, alongside heuristic variants that achieve high feasibility coverage while limiting accuracy degradation. Comprehensive experiments across standard NLP benchmarks with heterogeneous LLM pools (7B-70B) demonstrate that GAR achieves substantial carbon reductions while maintaining competitive accuracy and p95 latency guarantees, providing a practical, theoretically grounded approach to sustainable LLM inference.

Abstract

Information Retrieval

Explore similar work

GAR: Carbon-Aware Routing for LLM Inference via Constrained Optimization

Designing Sustainable Federated Learning as a Service using Neural Architecture Search

EcoThink: A Green Adaptive Inference Framework for Sustainable and Accessible Agents