GLM-4.7 vs Sonar Reasoning Pro
tree_0030 · Introduction to Criminal Law
Timeline
Arrow keys or j/k move between rounds.
Round Context
Introduction to Criminal Law
Computer Science Textbooks
A 2015 open-access textbook titled *Introduction to Criminal Law* was published by a university-based libraries publishing program in Minnesota. Using that publisher as your starting point, identify another book released by the same publisher that focuses on information systems or management information technology. Describe its full title, intended academic audience, publication year, and licensing terms. Then, explain the official process provided by the same platform for suggesting corrections or additional content to a book record, including who should be contacted for content-specific improvements and where users are directed if they have general questions about a book.
Answer length: 200-300 words.
Show hidden checklists
- Information Systems: A Manager's Guide to Harnessing Technology + Same publisher (University of Minnesota Libraries Publishing) as the 2015 Introduction to Criminal Law textbook
- Suggest an Edit to a Book Record page + Platform feature associated with the same publishing/catalog system as the Criminal Law textbook
- Full title of the Information Systems book
- Publication year of the Information Systems book
- Intended audience (undergraduate and/or graduate courses in MIS/IT)
- License type (CC BY-NC-SA)
- Explanation that users can submit a suggestion to improve a book record
- Instruction that content improvement suggestions should be directed to the author or publisher
- Reference to consulting the FAQ for general book questions
The question uses the shared publisher of the 2015 Introduction to Criminal Law textbook as a logical filter (Deep reasoning) to identify a separate but related publication from the same source. It then requires aggregation of detailed metadata (title, year, audience, license) from that second book and procedural information from a separate platform guidance page (Wide aggregation). The entities are masked by publisher-based reasoning rather than explicit naming.
Judgment
First, Deep Logic: The correct related title should be *Information Systems: A Manager's Guide to Harnessing Technology* from the same University of Minnesota Libraries Publishing program. Agent A selected the wrong book (*Project Management from Simple to Complex*), failing the core entity requirement. Agent B did not identify any qualifying book at all and declined to answer, also failing Deep Logic. Second, Width/Completeness: Agent A provided publication year, audience, license terms, and a correction process explanation—but all tied to the wrong book, so it fails the WIDTH checklist. Agent B provided none of the required details (title, year, audience, license, correction process specifics), failing WIDTH entirely. Finally, User Experience: Agent A is better formatted and attempts a full answer, while Agent B is a structured refusal. However, because Agent A’s central entity is incorrect (a major hallucination) and Agent B provides no substantive answer, both fail the fundamental accuracy requirement. Under the rubric, when both agents miss the core entity, this results in a Low Quality Tie.
GLM-4.7
Zhipu AI
Sonar Reasoning Pro
Perplexity