REVIEW 8 cited by
Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Do large language models (LLMs) know the law? These models are increasingly being used to augment legal practice, education, and research, yet their revolutionary potential is threatened by the presence of hallucinations -- textual output that is not consistent with legal facts. We present the first systematic evidence of these hallucinations, documenting LLMs' varying performance across jurisdictions, courts, time periods, and cases. Our work makes four key contributions. First, we develop a typology of legal hallucinations, providing a conceptual framework for future research in this area. Second, we find that legal hallucinations are alarmingly prevalent, occurring between 58% of the time with ChatGPT 4 and 88% with Llama 2, when these models are asked specific, verifiable questions about random federal court cases. Third, we illustrate that LLMs often fail to correct a user's incorrect legal assumptions in a contra-factual question setup. Fourth, we provide evidence that LLMs cannot always predict, or do not always know, when they are producing legal hallucinations. Taken together, our findings caution against the rapid and unsupervised integration of popular LLMs into legal tasks. Even experienced lawyers must remain wary of legal hallucinations, and the risks are highest for those who stand to benefit from LLMs the most -- pro se litigants or those without access to traditional legal resources.
Forward citations
Cited by 8 Pith papers
-
Think Through a Bottleneck: Hourglass Reasoning for Rigorous Induction
Strict stage isolation that passes only a compressed symbolic schema and rule between LLM calls improves few-shot inductive reasoning more than self-refinement or explicit verbalization alone.
-
Prompt Compression via Activation Aggregation
A learned weighted sum of intermediate-layer activations compresses an instruction prompt into a single patch vector that, injected at an early layer, recovers task accuracy within ~2% of the full prompt.
-
Automatically Evolving Prompt Guidelines for Task-Specific Optimization
AGOPS automatically evolves task-specific prompt guidelines from reference answers and reports recovering 15.5–81.7% of the performance lost to underspecified prompts.
-
Are the Hidden States Hiding Something? Testing the Limits of Factuality-Encoding Capabilities in LLMs
Hidden-state factuality probes trained on synthetic statements do not generalize to LLM-generated factual statements, despite reproducing prior results on original datasets.
-
A Reasoning-Focused Legal Retrieval Benchmark
Two hand-annotated legal RAG benchmarks with low query-passage lexical overlap show that existing retrievers struggle, and that structured legal reasoning query expansion improves retrieval.
-
Question Begets Question: Self-Evolving Curriculum for Reinforcement Fine-Tuning on Competition Mathematics
A self-evolving curriculum that retrains a language model on variants of problems it can mostly get right lifts AIME pass@1 from 5.6% to 16.5%, beating static augmentation under the same data budget.
-
Hybrid AI for Explainable and Accurate Conversational Agents in eGovernment
A chatbot architecture uses an LLM only to translate free text into typed inputs while a DCR rule graph controls the conversation and issues conclusions.
-
Elevating Legal LLM Responses: Harnessing Trainable Logical Structures and Semantic Knowledge with Legal Reasoning
LSIM combines reinforcement-learned fact-rule chains, a trainable DSSM retriever, and in-context learning to improve legal QA output over semantic-only RAG baselines by about 2 to 3 points on METEOR and ROUGE-1.
Discussion (0). Continue with ORCID to comment.