How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations

Ali Nemati5 days ago22 sec read30 views

Researchers conducted an audit of citation fabrication by large language models across four academic domains, finding hallucination rates vary widely depending on model and prompt framing. Key takeaway for content creators: using multi-model consensus and within-prompt repetition can significantly improve accuracy in detecting fabricated citations.

Read the full article at arXiv cs.CL (NLP)

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.

Comments

Open challenges in LLM research

The paper discusses six potential avenues for advancing large language model (LLM) research beyond scaling up existing models: improving performance o...The paper discusses six potential avenues for advancing large language model (LLM) research beyond scaling up existing models: improving performance on out-of-distribution data and reducing hallucinations through techniques like in-context learning; ...

Ali Nemati

AI & Machine Learning9 hours ago24 sec read

Animating Petascale Time-varying Data on Commodity Hardware with LLM-assisted Scripting

Scientists now have access to a user-friendly framework that enables them to create 3D animations of petascale time-varying datasets on standard works...Scientists now have access to a user-friendly framework that enables them to create 3D animations of petascale time-varying datasets on standard workstations using cloud-hosted repositories and an LLM-assisted conversational interface. This breakthro...

Ali Nemati

AI & Machine Learning2 days ago34 sec read

Building Next-Gen Agentic AI: A Complete Framework for Cognitive Blueprint Driven Runtime Agents with Memory Tools and Validation

The system architecture includes a cognitive blueprint defining agent roles and constraints, a tool registry for utility functions, and an LLM client ...The system architecture includes a cognitive blueprint defining agent roles and constraints, a tool registry for utility functions, and an LLM client for language model interactions. Memory management tracks conversation history and compresses it whe...

Ali Nemati

AI & Machine Learning5 days ago24 sec read

LWiAI Podcast #235 - Sonnet 4.6, Deep-thinking tokens, Anthropic vs Pentagon

The LWiAI Podcast #235 discusses recent advancements in AI models like Anthropic's Sonnet 4.6 and Google's Gemini 3.1 Pro, business moves by tech gian...The LWiAI Podcast #235 discusses recent advancements in AI models like Anthropic's Sonnet 4.6 and Google's Gemini 3.1 Pro, business moves by tech giants, infrastructure developments, and research on AI safety and policy. Key for content creators is t...

Ali Nemati

AI & Machine LearningFeb 2828 sec read

Memory Scaffolding Shapes LLM Inference: How Persistent Context Changes What AI Builds

The article demonstrates how persistent memory scaffolding significantly alters Large Language Model (LLM) outputs and reasoning processes, even when ...The article demonstrates how persistent memory scaffolding significantly alters Large Language Model (LLM) outputs and reasoning processes, even when using identical prompts and models. This technique injects context that shapes architectural density...

Ali Nemati

How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations

Related Articles

Open challenges in LLM research

Animating Petascale Time-varying Data on Commodity Hardware with LLM-assisted Scripting

Building Next-Gen Agentic AI: A Complete Framework for Cognitive Blueprint Driven Runtime Agents with Memory Tools and Validation

LWiAI Podcast #235 - Sonnet 4.6, Deep-thinking tokens, Anthropic vs Pentagon

Memory Scaffolding Shapes LLM Inference: How Persistent Context Changes What AI Builds