Cover Image

Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Garg, Spandan, Nitin, Vikram, Huang, Yufan
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence Software Engineering
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Debug2Fix: Can Interactive Debugging Help Coding Agents Fix More Bugs?
by: Garg, Spandan, et al.
Published: (2026)

Saving SWE-Bench: A Benchmark Mutation Approach for Realistic Agent Evaluation
by: Garg, Spandan, et al.
Published: (2025)

The SWE-Bench Illusion: When State-of-the-Art LLMs Remember Instead of Reason
by: Liang, Shanchao, et al.
Published: (2025)

Can LLMs Replace Humans During Code Chunking?
by: Glasz, Christopher, et al.
Published: (2025)

PerfBench: Can Agents Resolve Real-World Performance Bugs?
by: Garg, Spandan, et al.
Published: (2025)

RAPGen: An Approach for Fixing Code Inefficiencies in Zero-Shot
by: Garg, Spandan, et al.
Published: (2023)

Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering
by: Wang, Ruiqi, et al.
Published: (2025)

Agentic Scientific Simulation: Execution-Grounded Model Construction and Reconstruction
by: Lie, Knut-Andreas, et al.
Published: (2026)

Project Prometheus: Bridging the Intent Gap in Agentic Program Repair via Reverse-Engineered Executable Specifications
by: Wang, Yongchao, et al.
Published: (2026)

The Semi-Executable Stack: Agentic Software Engineering and the Expanding Scope of SE
by: Feldt, Robert, et al.
Published: (2026)

Smaller = Weaker? Benchmarking Robustness of Quantized LLMs in Code Generation
by: Fang, Sen, et al.
Published: (2025)

Teaching LLMs to Learn Tool Trialing and Execution through Environment Interaction
by: Gao, Xingjie, et al.
Published: (2026)

Agentic Frameworks for Reasoning Tasks: An Empirical Study
by: Rasheed, Zeeshan, et al.
Published: (2026)

How Do LLMs Fail In Agentic Scenarios? A Qualitative Analysis of Success and Failure Scenarios of Various LLMs in Agentic Simulations
by: Roig, JV
Published: (2025)

Runtime-Structured Task Decomposition for Agentic Coding Systems
by: Asthana, Shubhi, et al.
Published: (2026)

Integrating Symbolic Execution into the Fine-Tuning of Code-Generating LLMs
by: Sakharova, Marina, et al.
Published: (2025)

Securing the AI Frontier: Urgent Ethical and Regulatory Imperatives for AI-Driven Cybersecurity
by: Kulothungan, Vikram
Published: (2025)

Evaluating LLMs for Visualization Tasks
by: Khan, Saadiq Rauf, et al.
Published: (2025)

Tree-of-Code: A Hybrid Approach for Robust Complex Task Planning and Execution
by: Ni, Ziyi, et al.
Published: (2024)

Generating Structured Plan Representation of Procedures with LLMs
by: Garg, Deepeka, et al.
Published: (2025)

RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts
by: Weidener, Lukas, et al.
Published: (2026)

DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use
by: Chen, Aili, et al.
Published: (2026)

SWE-Hub: A Unified Production System for Scalable, Executable Software Engineering Tasks
by: Zeng, Yucheng, et al.
Published: (2026)

Can LLMs Generate User Stories and Assess Their Quality?
by: Quattrocchi, Giovanni, et al.
Published: (2025)

Process-Supervised Reinforcement Learning for Code Generation
by: Ye, Yufan, et al.
Published: (2025)

Capture the Flags: Family-Based Evaluation of Agentic LLMs via Semantics-Preserving Transformations
by: Honarvar, Shahin, et al.
Published: (2026)

Benchmarks are Not Enough: RAMP for Runtime Assessing of Agentic Models in Production Systems
by: Ouyang, Yipeng, et al.
Published: (2026)

RA-Gen: A Controllable Code Generation Framework Using ReAct for Multi-Agent Task Execution
by: Liu, Aofan, et al.
Published: (2025)

Ambiguity Detection and Elimination in Automated Executable Process Modeling
by: Matei, Ion, et al.
Published: (2026)

Tree-of-Code: A Tree-Structured Exploring Framework for End-to-End Code Generation and Execution in Complex Task Handling
by: Ni, Ziyi, et al.
Published: (2024)

SHERPA: A Model-Driven Framework for Large Language Model Execution
by: Chen, Boqi, et al.
Published: (2025)

DeepCode: Open Agentic Coding
by: Li, Zongwei, et al.
Published: (2025)

EduBot -- Can LLMs Solve Personalized Learning and Programming Assignments?
by: Wang, Yibin, et al.
Published: (2025)

Constraint-Guided Multi-Agent Decompilation for Executable Binary Recovery
by: Zhang, Yifan, et al.
Published: (2026)

Pragmos: A Process Agentic Modeling System
by: Hernández-Ávalos, Pedro-Aarón, et al.
Published: (2026)

DynamicsLLM: a Dynamic Analysis-based Tool for Generating Intelligent Execution Traces Using LLMs to Detect Android Behavioural Code Smells
by: Cherief, Houcine Abdelkader, et al.
Published: (2026)

How Robustly do LLMs Understand Execution Semantics?
by: Spiess, Claudio, et al.
Published: (2026)

Treefix: Enabling Execution with a Tree of Prefixes
by: Souza, Beatriz, et al.
Published: (2025)

PEFA-AI: Advancing Open-source LLMs for RTL generation using Progressive Error Feedback Agentic-AI
by: Narayanan, Athma, et al.
Published: (2025)

Insights from Benchmarking Frontier Language Models on Web App Code Generation
by: Cui, Yi
Published: (2024)