Knowledge Tree
Relations (1)
related 2.81 — strongly supporting 6 facts
GPT-4 and GPT-3.5 are related as they are both frequently evaluated and compared as state-of-the-art large language models in various research contexts, including task performance benchmarking [1], instruction adherence [2], knowledge graph construction [3], and KBQA tasks [4]. They are also both subject to shared vulnerabilities, such as performance degradation under adversarial attacks in time series forecasting [5], and are often contrasted based on their relative capabilities [6].
Facts (6)
Sources
The construction and refined extraction techniques of knowledge ... nature.com 2 facts
claimGPT-3.5 is a language model capable of natural language understanding and generation, though it performs with more limitations compared to GPT-4.
claimThe study compares a fine-tuned DeepSeek-R1 70B LoRA model against baseline models including the original DeepSeek-R1 70B, GPT-4, GPT-3.5, and LLaMA3 70B to assess task performance improvements.
Large Language Models Meet Knowledge Graphs for Question ... arxiv.org 1 fact
referenceThe ODA method, proposed by Sun et al. in 2024, uses ODA-based knowledge graph retrieval with GPT-4 and GPT-3.5 models to perform KBQA tasks, evaluated using Hits@1 and Acc metrics on the QALD10-en dataset.
Track: Poster Session 3 - aistats 2026 virtual.aistats.org 1 fact
claimAdversarial attacks on Large Language Models (LLMs) for time series forecasting lead to more severe performance degradation than random noise across models including LLMTime with GPT-3.5, GPT-4, LLaMa, Mistral, TimeGPT, and TimeLLM.
Building Trustworthy NeuroSymbolic AI Systems - arXiv arxiv.org 1 fact
claimGPT-3.5, Claude, and GPT-4.0 adhere more closely to instructions than LLama2 (Touvron et al. 2023), Vicuna (Chiang et al. 2023), and Falcon (Penedo et al. 2023).
Unknown source 1 fact
claimThe research paper 'Towards the Automation of Knowledge Graph Construction Using ...' explores the semi-automatic and automatic construction of knowledge graphs using state-of-the-art large language models including Mixtral 8x22B Instruct v0.1, GPT-4o, and GPT-3.5.