PulseAugur
EN
LIVE 19:02:06

New benchmarks reveal LLM limitations in code generation and translation

By PulseAugur Editorial · [2 sources] ·

Two new benchmarks, E2E-SWE and ORCA, have been introduced to evaluate the capabilities of large language models (LLMs) in complex coding tasks. E2E-SWE focuses on LLMs' ability to generate entire software repositories from scratch, testing 13 frontier models across 11 programming languages with results showing significant variation in performance. ORCA, on the other hand, assesses LLMs' proficiency in translating code between data science libraries, revealing that even advanced models like Claude Opus-4.6 struggle with this task, prompting the development of an intent-augmented translation method. AI

IMPACT These benchmarks highlight current LLM limitations in complex software engineering tasks, indicating areas for future research and development in AI coding capabilities.

RANK_REASON The cluster consists of two academic papers introducing new benchmarks for evaluating LLMs.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New benchmarks reveal LLM limitations in code generation and translation

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
The cluster consists of two academic papers introducing new benchmarks for evaluating LLMs.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
14 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. arXiv cs.AI TIER_1 English(EN) · Hantian Ding, Chloe Bi, Jiacheng Zhu, John Yang, Matt Deitke, Pengcheng Yin, Zijian Wang, Rui Hou ·

    E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch

    arXiv:2609.38335v1 Announce Type: cross Abstract: Coding agents powered by large language models (LLMs) are evolving from making localized code changes to developing complete software repositories. However, evaluating repository-scale generation remains challenging: tasks must de…

  2. arXiv cs.AI TIER_1 English(EN) · Xiaolong Li, Jinyang Li, Bowen Qin, Ge Qu, Nan Huo, Xiaohan Xu, Shipei Lin, Reynold Cheng ·

    ORCA: Evaluating LLMs on Data Science Code Translation

    arXiv:2609.30749v1 Announce Type: new Abstract: Data Science Code Translation (DSCT) is the process of converting code between data science libraries while preserving functional equivalence and enabling interoperability across data science ecosystems. While Large Language Models …

RELATED ENTITIES

RELATED TOPICS