ADAbhiyan Dhakal
Menu

Questions, methods, evidence

Research

I am interested in making computing more efficient and predictable under resource constraints. Controlled, reproducible experiments are the method; resource behavior is the subject.

Current work

Current systems research on Linux memory pressure, responsiveness, and the signals available before a machine becomes unusable under load.

ongoing2026-present

Linux memory pressure and PSI signal validation

Broader goal

How can memory-management policy preserve responsiveness when RAM is constrained?

Evidence to date

Controlled reruns show PSI before material swap growth under the tested memory-pressure workloads. A corrected, phase-aligned responsiveness canary now records pressure-related response and service-time degradation, while compression-sensitivity runs distinguish structured from random page content.

Active policy evaluation

  1. Run disk-swap, static-zswap, and static-zram controls under identical structured and random background pressure.

  2. Use sustained PSI-some from a bounded background cgroup to activate zram once, while retaining PSI-full as diagnostic telemetry.

  3. Compare foreground response and service-time tails, deadline misses, swap growth, backend occupancy, faults, reclaim, and no-trigger cases against the static controls.

Status and next stage

Baseline signal validation and compression-sensitivity work are complete. The cgroup-scoped sustained-PSI guided-zram comparison is underway. The current evidence does not establish that guided zram improves responsiveness or outperforms static swap backends; that is the purpose of the active policy screen.

Previous applied research

Nepali legal information retrieval in a low-resource language setting.

The 2025 and 2026 ICAIL papers form a continuing empirical track: first examining retrieval-augmented legal analysis in Nepal, then isolating and comparing embedding models as a retrieval decision. This work provided experience with imperfect data, evaluation design, publication, and presenting results.

Publications

2026
conference
ICAIL 2026
Short paper - presented 9 June 2026

An Empirical Comparison of Embedding Models for Nepali Legal Document Retrieval

Abhiyan Dhakal, Kausik Paudel, Pranish Kafle, Sugat Sujakhu, Prakash Poudyal, Anita Jadhari

An empirical study of embedding choices for retrieving Nepali legal documents in a low-resource setting.

Study contribution: Compares retrieval behavior rather than treating embedding choice as an implementation detail, with attention to the constraints of Nepali legal text.

Official programme
2025
preprint
arXiv:2509.15292; National Conference on Computer Innovations
Preprint

An Artificial Intelligence Driven Semantic Similarity-Based Pipeline for Rapid Literature

Abhiyan Dhakal, Kausik Paudel, Sanjog Sigdel

A lightweight pipeline that generates search terms, retrieves open-access papers, ranks them with transformer embeddings, and applies statistical thresholding.

Study contribution: Evaluates three embedding models and documents a practical preliminary-review workflow without claiming ground-truth relevance labels that were not collected.

arXivDOICode companion
2025
conference
Proceedings of ICAIL 2025, pp. 498-499
Published

Feasibility of Artificial Intelligence Driven Analysis in the Context of Nepalese Legal System

Abhiyan Dhakal, Sugat Sujakhu, Pranish Kafle, Kausik Paudel, Prakash Poudyal

A retrieval-augmented legal analysis pipeline designed for access to information in the Nepalese legal context.

Study contribution: Processes legal documents for retrieval, expands user queries, and evaluates rule-recall, rhetorical-understanding, and interpretation-oriented questions.

DOIDBLP