HONG KONG, Nov. 18, 2026 /PRNewswire/ -- In the latest Times Higher Education Interdisciplinary Science Rankings 2027 in association with Schmidt Science Fellows, City University of Hong Kong (CityUHK) was ranked 16th globally, 5th in Asia and 2nd in Hong Kong.
The number of universities participating in this year's ranking reached a record high, with 1,058 institutions ranked worldwide. Hong Kong's higher education sector delivered a standout performance, with three universities—including CityUHK—breaking into the global top 20 and two securing spots in the Asian top 5, highlighting the city's strong momentum in interdisciplinary research. Ranking 16th globally, CityUHK also achieved a score of 93 out of 100 in the Inputs category, maintaining a strong performance that reflects the University's sustained commitment to investing resources in and advancing interdisciplinary research.
The Interdisciplinary Science Rankings assess universities' performance across three stages in the life cycle of research projects: "Inputs" (funding), "Process" (measures of success, physical facilities, administrative support, and the promotion process) and "Outputs" (number and proportion of publications, research quality and reputation).
A strong advocate for interdisciplinary scientific research, CityUHK has established multiple institutes and colleges in recent years, including the CityUHK Academy of Innovation, the Institute of Digital Medicine, the Hong Kong Institute of AI for Science, the Institute for Materials Innovation and the Institute of Future Learning, so as to accelerate solutions to pressing global issues. The University has also launched a new Bachelor of Laws and Bachelor of Science in Computer Science (Stream: Artificial Intelligence) double degree programme, which will admit its first cohort in the 2027/28 academic year. The programme aims to nurture professionals with expertise in both law and AI to address the legal challenges arising from rapid technological advancement.
CityUHK's academic excellence continues to gain international recognition. Recently, 31 CityUHK scholars were named to the 2026 Highly Ranked Scholars lifetime list released by ScholarGPS, a US-based scholarly data platform. CityUHK has the largest number of listed scholars among Hong Kong institutions and ranks first among all universities and institutions nationwide. In the "World's Top 2% Scientists" 2026 by Stanford University, released last month, CityUHK had 263 scholars named on the list, based on their career-long citation impact.
Furthermore, in the QS World University Rankings 2027, CityUHK ranked first in Asia and second globally, trailing only Harvard University in the indicator of Citations per Faculty.
** This press release is distributed by PR Newswire through automated distribution system, for which the client assumes full responsibility. **
CityUHK ranks 16th globally and 5th in Asia in THE Interdisciplinary Science Rankings 2027
Top-performing model achieves just 26.7% accuracy, revealing a significant gap in autonomous rule discovery
SINGAPORE, Nov. 18, 2026 /PRNewswire/ -- As AI agents evolve from answering questions to completing complex tasks, a fundamental challenge remains: Can they solve problems when instructions are absent and rules must be discovered entirely from visual clues?
Einsia AI's Navers Lab today announced Visual Cipher Bench, a new benchmark evaluating long-horizon visual reasoning in AI agents, focusing on rule discovery, logical planning, and spatial operations.
Across eight frontier models and 29 reasoning configurations, the benchmark recorded 3,480 attempts. No configuration achieved accuracy above 30%. The top performer, GPT-6 Astra (max), scored 26.7%. DeepSeek V4.1 Flash, Qwen 3.8 Max, and Kimi K3.1 Pro failed to solve any puzzles.
The findings expose a critical gap between executing known instructions and independently discovering unknown rules. While spatial reasoning proved relatively manageable, accuracy on rule induction was just 0.6%.
Testing AI Beyond Conventional Visual Question Answering
Unlike conventional visual benchmarks, which typically provide explicit questions and expected answers, Visual Cipher Bench tests scenarios where instructions are hidden in images, with no explicit rules or immediate correctness feedback.
The benchmark comprises 40 original visual puzzles represented by 67 black-and-white illustrations. Agents receive one instruction: examine the images and write the final answer to a file. They must independently identify examples, understand the target problem, infer the required output, and solve each task.
The puzzles cover four categories:
- Spatial Operations: Folding, rotation, projection, and geometric transformations.
- Logical Planning: Extracting visual constraints and finding valid solutions.
- Rule Induction: Discovering hidden patterns from limited examples.
- Knowledge Decoding: Interpreting Morse code, Braille, and word puzzles.
Frontier Models Struggle with Autonomous Rule Discovery
Performance varied significantly across the eight evaluated models:
- GPT-6 Astra (max): 26.7%, averaging 10.7 correct answers per 40-puzzle run.
- GPT-6.1 Sol (max) and Claude Opus 5.5 (xhigh): Both scored 21.7%.
- GPT-6 Luna and Claude Sonnet 5.5: Both scored 0.8%, solving one puzzle each across 120 attempts.
- DeepSeek V4.1 Flash, Qwen 3.8 Max, and Kimi K3.1 Pro: Zero correct answers.
Across all 3,480 attempts, accuracy by category was:
- Spatial Operations: 22.8%
- Knowledge Decoding: 3.3%
- Logical Planning: 2.4%
- Rule Induction: 0.6%
Models solved only five of 870 rule-induction attempts. Even GPT-6 Astra (max) achieved 80% on spatial operations, while no configuration exceeded 13.3% in the other three categories. Researchers also found that some correct answers resulted from thematic guesses rather than systematic rule discovery.
Providing Rules Raises Accuracy from 12.2% to 51.1%
To distinguish rule discovery from rule execution, researchers conducted an additional experiment on 10 puzzles using GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna.
When provided with the core rules—but not intermediate reasoning steps or answers—overall accuracy increased from 12.2% to 51.1%. Rule-induction accuracy rose from 0% to 85.2%.
GPT-6 Astra improved from 23.3% to 70.0%, while GPT-6.1 Sol increased from 13.3% to 66.7%. Timeout occurrences also fell from 51 to 29 across 90 attempts.
A control experiment providing additional information about task-state representation, but not the core rules, achieved only 16.7% accuracy. Even with rule hints, logical planning accuracy remained at 22.2%.
Why Do AI Agents Fail?
Analysis of failure trajectories revealed two recurring patterns.
Premature conclusions: Agents often adopted rules that explained only some examples without validating them against all available evidence. Some submitted answers despite acknowledging incomplete understanding.
Inefficient search: Agents frequently relied on extensive tool calls and exhaustive combinations instead of using visual constraints to narrow the search space, eventually timing out.
More effective reasoning involved generating multiple hypotheses, testing them against examples, eliminating inconsistent rules, and applying the remaining valid hypothesis.
From Following Instructions to Discovering Rules
In software development, agents typically receive explicit requirements and can rely on test results to evaluate their solutions. Real-world problems often provide neither.
Agents must observe their environment, identify relevant signals, formulate hypotheses, and validate their understanding without explicit instructions or immediate feedback.
Visual Cipher Bench provides a systematic evaluation of this capability gap. Its findings suggest that independently discovering, validating, and transferring hidden rules remains a significant challenge for frontier AI agents —a capability increasingly important as AI moves toward more autonomous problem-solving.
Research and Resources
** This press release is distributed by PR Newswire through automated distribution system, for which the client assumes full responsibility. **
Einsia AI Launches Visual Cipher Bench to Evaluate Long-Horizon Visual Reasoning in AI Agents
Einsia AI Launches Visual Cipher Bench to Evaluate Long-Horizon Visual Reasoning in AI Agents
Einsia AI Launches Visual Cipher Bench to Evaluate Long-Horizon Visual Reasoning in AI Agents
Einsia AI Launches Visual Cipher Bench to Evaluate Long-Horizon Visual Reasoning in AI Agents