Research Publications
Research Publication
SWE-MERA: A Dynamic Benchmark for Agenticly Evaluating Large Language Models on Software Engineering Tasks
Conference on Empirical Methods in Natural Language Processing: System Demonstrations (EMNLP)Jan 1, 20250 OpenAlex citationsMatched to OpenAlex
DOI: 10.18653/v1/2025.emnlp-demos.30
Publication summary
Matched to OpenAlexNo abstract is available yet. The publication is shown because it is linked to an approved public researcher profile.
Software Engineering ResearchTopic ModelingSoftware Engineering Techniques and Practiceslarge language modelssoftware engineering tasksbenchmarksoftware agentscode generation evaluation
Authors
Related publications
Connected through authors and topics already present in the Research Hub.
Beyond Attention: Module-Level Refusal Directions in Mamba-Based Hybrid Language ModelsLecture notes in computer science · 2026 · 0 OpenAlex citationsSame author: Valentin MalykhA Reproducible Microscopy Acquisition Pipeline and Microscopy-Aware Attention for Classification, Detection, and Segmentation of MicroorganismsMachine Learning and Knowledge Extraction · 2026 · 0 OpenAlex citationsSame author: Valentin MalykhHybrid Non-local Blocks for Streptococci Segmentation in Microscopic ImagesLecture notes in networks and systems · 2026 · 0 OpenAlex citationsSame author: Valentin MalykhFlow-Guided Neural Pruning: Signal-Flow Framework for Multi-Architecture Model CompressionMachine Learning and Knowledge Extraction · 2026 · 0 OpenAlex citationsSame author: Valentin Malykh