Lakshmi Pedapudi

—

Unclaimed · AI-structured profileOpenAlex

Early-career applied AI researcher pairing agricultural computer vision with speech processing for farm settings

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Output scoring is unavailable in this snapshot: only h-index 1, i10-index 0 and year-level work counts exist, without a full publication list or venue detail. This axis would measure total indexed works and their per-year distribution.

Impact—

Impact cannot yet be scored: citation indicators are minimal (h-index 1, i10-index 0) and no per-year citation series is included in the snapshot. This axis would measure total citations, citations per paper, and uptake of the reviewed pipelines.

Focus0/100

Focus scores 0, reflecting breadth rather than a single deep lane: the OpenAlex field distribution splits exactly 50/50 between applied agricultural computer vision and speech recognition, and topic clustering spreads the record across five groups — industrial vision systems and defect detection (5 works), smart agriculture and AI (3), speech recognition and synthesis (2), spectroscopy and chemometric analysis (2), and speech and audio processing (2). The two reviewed 2026 papers sit one in each domain.

Collaboration48/100

Collaboration scores 48, built on 10 distinct coauthors within the recent three-year window. The recurring co-author pool includes Naga Ganesh, Chandrashekar M. S., Aakash Singh, Vineet Singh and Sanyam Singh, pointing to a small stable team. Both reviewed 2026 papers list Pedapudi in a middle authorship position without corresponding-author status, so the current profile is a contributing team member rather than a lead.

Activity61/100

Activity scores 61. The index trail counts 2 works in 2021, 1 in 2022 and 2 in 2024, then 5 in 2026 — the strongest year on record and the source of both reviewed arXiv preprints. Recent three-year output stands at 7 works, so the curve is pointing up even though absolute volume remains that of an early-stage researcher.

Reproducibility78/100

Reproducibility is the strongest axis at 78, the mean across the 2 papers Evolve has assessed. The agriculture voice pipeline earns a high readiness rating on a well-released, no-finetune setup, while the crop-diagnosis vision pipeline is rated medium because panel-made labels cap what can be independently verified. Pairing models with reusable setups is already a consistent habit here.

Research summary

Lakshmi Pedapudi is an early-stage applied AI researcher whose indexed work divides evenly between two domains: OpenAlex field distribution assigns 0.5 to applied computer vision for agriculture and 0.5 to speech processing for automatic speech recognition. The two 2026 arXiv preprints reviewed by Evolve illustrate the split. The first rebuilds crop disease and pest diagnosis as a configurable three-stage vision pipeline in which DaViT-Base outperforms Plantix and a fine-tuned vision-language model, with MobileNetV3 and Qwen3-VL in the comparison set; Evolve rates its reproducibility medium, noting that panel-made labels cap the evidence. The second proposes a model-agnostic and language-agnostic voice pipeline for noisy multilingual farm audio, cutting word error rate 16-23% on cloud ASR without fine-tuning, with most of the gain traced to speaker selection; Evolve calls it well-released, credible work and rates reproducibility high. Output volume is modest but current: 7 works in the last three years, 5 of them in 2026, within an OpenAlex trail that starts in 2021 with 2 works. Citation indicators remain thin — h-index 1, i10-index 0 — so the impact axis is left unscored. Ten distinct coauthors appear in the recent window, and both reviewed papers place Pedapudi in a middle, non-corresponding authorship position. Reproducibility is the standout axis, averaging 78 across the two assessed papers.

Collaboration pattern

Pedapudi works inside a small, recurring team: 10 distinct coauthors appear in the recent three-year window, with Naga Ganesh, Chandrashekar M. S., Aakash Singh, Vineet Singh and Sanyam Singh recurring most often. Neither reviewed 2026 paper lists them as corresponding author, and both place them in a middle authorship slot — a contributor profile, consistent with the collaboration score of 48.

Research keywords

crop disease detectionpest classificationvision-language modelsmodel-agnostic pipelinesautomatic speech recognitionspeaker diarizationlow-resource languagesIndic languagesspeech enhancementagricultural AIchemometric analysis

Highlights

Configurable three-stage vision pipeline tops Plantix on crop diagnosis
A production-data study rebuilds crop disease and pest diagnosis as three configurable stages; DaViT-Base outperforms Plantix and a fine-tuned vision-language model, with MobileNetV3 and Qwen3-VL in the comparison set. Evolve notes the evidence is capped by panel-made labels and rates reproducibility medium.
No-finetune voice pipeline cuts noisy farm-audio WER 16-23%
A model-agnostic and language-agnostic ASR pipeline for noisy multilingual agricultural audio reduces word error rate by 16-23% on cloud systems, with most of the gain coming from speaker selection. Evolve calls it well-released, credible work with high reproducibility readiness.
Reproducibility is the strongest signal in this profile
Across the 2 papers Evolve has assessed, reproducibility readiness averages 78: the agriculture voice pipeline is rated high on a well-released no-finetune design, while the crop-diagnosis pipeline is rated medium because panel-made labels cap what can be independently verified.

Timeline

2021Earliest indexed activity in the OpenAlex record: 2 works published.
2022Indexed output thins to a single work.
2024Output resumes at 2 indexed works after the quiet 2023 in the provided counts.
2026Most productive year on record: 5 works, including both Evolve-reviewed arXiv preprints — the crop-diagnosis vision pipeline and the agriculture voice pipeline.

Papers in the Evolve catalog

Frequent co-authors

Naga Ganesh×2 Chandrashekar M S×2 Naga Ganesh×1 Chandrashekar M S×1 Lakshmi Pedapudi×1 Aakash Singh×1 Vineet Singh×1 Aakash Singh×1 Sanyam Singh×1 Vineet Singh×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Bibliometrics come from OpenAlex only; no ORCID, institution, or country is verified, so this profile is name-resolved from public bibliographic data.
Output and impact axes are unscored: the record holds only h-index 1, i10-index 0 and year-level counts, which is too thin to rank citation impact.
Only the 2 listed papers have been reviewed by Evolve; topic and field distributions are algorithmic OpenAlex clusterings, not curatorial judgments.
The frequent-coauthor list contains duplicate entries and the researcher's own name; treat co-author signals as a small sample.