No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Behavioral Emotion Analysis Model for Large Language Models
... SocraSynth's adversarial component promotes the exploration of diverse perspectives, while its collaborative component fosters rigorous reasoning to reach well-reasoned conclusions. This synergy has yielded measurable gains beyond healthcare and bias mitigation, extending to geopolitical analysis (Chang, 2023b), corporate planning (Tsao, 2023), investment banking (Chang, 2024b), and emotional behavior modeling (Chang, 2024a). These results demonstrate SocraSynth's effectiveness in mitigating LLM limitations and achieving substantial performance improvements across various applications, highlighting its potential for advancing towards AGI's generalized problemsolving capabilities. ...
... Our empirical validation demonstrates EVINCE's effectiveness in improving prediction accuracy across various domains, notably achieving a 5% improvement in medical diagnosis tasks. The framework has also shown promise in identifying biases in news articles (Chang, 2024c), showcasing its potential for broader applications in fields such as geopolitical analysis (Chang, 2023b) corporate planning (Tsao, 2023), and emotional behavior modeling (Chang, 2024a). ...
This paper introduces EVINCE (Entropy and Variation IN Conditional Exchanges), a dialogue framework advancing Artificial General Intelligence (AGI) by enhancing versatility, adaptivity, and reasoning in large language models (LLMs). Leveraging adversarial debate and a novel dual entropy theory, EVINCE improves prediction accuracy, robustness, and stability in LLMs by integrating statistical modeling, information theory, and machine learning to balance diverse perspective exploration with strong prior exploitation. The framework's effectiveness is demonstrated through consistent convergence of information-theoretic metrics, particularly improved mutual information, fostering productive LLM collaboration. We apply EVINCE to healthcare, showing improved disease diagnosis, and discuss its broader implications for decision-making across domains. This work provides theoretical foundations and empirical validation for EVINCE, paving the way for advancements in LLM collaboration and AGI development.
... SocraSynth, our previous work [25] presented in Chapter 6, addresses LLM limitations through structured multi-agent dialogues. By leveraging both adversarial and collaborative interactions between LLMs, SocraSynth demonstrates quantifiable improvements across various domains, including healthcare [34], news analysis [32], geopolitical analysis [28], corporate planning [138], and emotional behavior modeling [23,27]. These results highlight SocraSynth's potential for advancing towards AGI's generalized problem-solving capabilities. ...
This booklet, "Unlocking the Wisdom of LLM Collaborative Intelligence," serves as an introduction to the full-length work, "The Path to Artificial General Intelligence." Through ten carefully crafted aphorisms, it distills the core insights and guiding principles that underpin the broader exploration of AI’s future through LLM Collaborative Intelligence (LCI). The author presents this framework as a promising pathway toward achieving artificial general intelligence (AGI).
As the global AI community races toward AGI, key figures like Yann LeCun argue that LLMs alone are insufficient for reaching AGI. LeCun, a pioneer in deep learning, suggests that text-based models are inherently limited due to their lack of persistent memory, physical interaction, and planning abilities. He insists that true intelligence requires direct interaction with the physical world—such as through robots or sensors—arguing that language-only systems cannot achieve the grounding needed for general intelligence.
In contrast, this book proposes an alternative view: LCI demonstrates that the exchange of information, contextual adaptation, and collaborative dialogue can overcome many of the limitations LeCun highlights. The key is not to rely on isolated LLMs but to leverage their synergy through structured communication, expanding their capacities beyond any individual model.
... This is achieved through a dynamic protocol that adaptively adjusts the "contentiousness" level of the debate, encouraging models to initially explore a wide range of perspectives and rigorously assess the quality of arguments. By leveraging both adversarial and collaborative interactions between LLMs, SocraSynth demonstrates quantifiable improvements across various domains, including healthcare Chang & et al. (2023), sales planning Tsao (2023), and emotional behavior modeling Chang (2024). These results highlight the potential for advancing towards AGI's generalized problem-solving capabilities. ...
Misdiagnosis is a significant issue in healthcare, leading to harmful consequences for patients. The propagation of mislabeled data through machine learning models into clinical practice is unacceptable. This paper proposes EVINCE, a system designed to 1) improve diagnosis accuracy and 2) rectify misdiagnoses and minimize training data errors. EVINCE stands for Entropy Variation through Information Duality with Equal Competence, leveraging this novel theory to optimize the diagnostic process using multiple Large Language Models (LLMs) in a struc-tured debate framework. Our empirical study verifies EVINCE to be effective in achieving its design goals.
This study introduces SocraHealth, an innovative method using Large Language Models (LLMs) for medical diagnostics. By engaging LLM-based agents in structured debates, SocraHealth not only refines diagnoses but also corrects historical record inaccuracies, utilizing patient data effectively. The case study, featuring GPT-4 and Bard across two experiments, showcases this approach's success in producing logical, hallucination-free debates. Demonstrating a significant advancement over traditional diagnostic techniques, SocraHealth highlights the transformative power of LLMs in healthcare, especially in enhancing diagnostic accuracy and rectifying past diagnostic errors.
This study explores the architectural advancements of large language models (LLMs), with a particular focus on the GPT-4 model. We begin with a thorough analysis of GPT-4’s distinctive features, including its polydisciplinary and polymodal data representation, the balanced approach in its algorithmic training, and the synergistic blend of human-driven insights with data-centric learning processes.
Building upon these insights, we introduce SocraSynth, a {\em reasoning layer} thoughtfully crafted to augment knowledge discovery and bolster analytical reasoning across an ensemble of LLMs. SocraSynth is designed to facilitate a generative process through multi-agent analytical discussions, followed by the evaluation of the resultant arguments for their ``reasonableness.'' This approach significantly enhances interdisciplinary information discovery and complex reasoning, strategically addressing major challenges faced by LLMs, such as the production of contextually inaccurate responses (hallucinations) and entrenched statistical biases. Implementing SocraSynth across various application domains marks a significant advancement in overcoming the limitations of current LLMs, paving the way for more reliable and sophisticated AI-driven analytical tools.
The CoCoMo model proposes a computational solution to the challenge of incorporating ethical and emotional intelligence considerations into AI systems, with the aim of creating AI agents that combine knowledge with compassion. To achieve this goal, Co-CoMo prioritizes fairness, beneficence, non-maleficence, empathy, adaptability, transparency, and critical and exploratory thinking abilities. The model employs consciousness modeling, reinforcement learning, and prompt template formulation to support these desired traits. By incorporating ethical and emotional intelligence considerations, a generative AI model can potentially lead to improved fairness, reduced toxicity, and increased reliability.
This paper presents a systematic approach to using the Socratic method in developing prompt templates that effectively interact with large language models, including GPT-3. Various methods are examined, and those that yield precise answers and justifications while fostering creativity and imagination to enhance creative writing are identified. Techniques such as definition, elenchus, dialectic, maieutics, generalization, and counterfactual reasoning are discussed for their application in engineering prompt templates and their connections to inductive, deductive, and abductive reasoning. Through examples, the effectiveness of these dialogue and reasoning methods is demonstrated. An interesting observation is made that when the task's goal and user intent are conveyed to GPT-3 via ChatGPT before the start of a dialogue, the large language model seems to connect to the external context expressed in the intent and perform more effectively. Index Terms-large language model, natural language processing , prompting, the Socratic method.
We present a new supervised learning procedure for systems composed of many separate networks, each of which learns to handle a subset of the complete set of training cases. The new procedure can be viewed either as a modular version of a multilayer supervised network, or as an associative version of competitive learning. It therefore provides a new link between these two apparently different approaches. We demonstrate that the learning procedure divides up a vowel discrimination task into appropriate subtasks, each of which can be solved by a very simple expert network.
Preparation requires technical research and development, as well as adaptive, proactive governance
This research addresses the dual objectives of improving the quality and reducing bias in Wikipedia and news articles. Quality is evaluated in terms of breadth, depth, accuracy, and neutrality, reflecting both the soundness of the content and the authority of references. We conceptualize bias as any tilt or slant in the information presented. Our methodology employs multiple Large Language Models (LLMs) in a novel way to appraise and refine articles from different standpoints. One LLM acts as an advocate for the article's current state, promoting its strengths and integrity. Concurrently, other LLMs scrutinize and challenge the article, applying defined metrics for quality and impartiality. This dialectical approach culminates in a synthesized enhancement that consolidates diverse insights, thus advancing the article's quality by diminishing bias. Our empirical findings substantiate the effectiveness of this technique in concurrently advancing the neutrality and caliber of content.
Bagging predictors is a method for generating multiple versions of a predictor and using these to get an aggregated predictor. The aggregation averages over the versions when predicting a numerical outcome and does a plurality vote when predicting a class. The multiple versions are formed by making bootstrap replicates of the learning set and using these as new learning sets. Tests on real and simulated data sets using classification and regression trees and subset selection in linear regression show that bagging can give substantial gains in accuracy. The vital element is the instability of the prediction method. If perturbing the learning set can cause significant changes in the predictor constructed, then bagging can improve accuracy.
"Foundations of Large-Scale Multimedia Information Management and Retrieval: Mathematics of Perception" pioneers the data-driven approach in machine learning, highlighting the author's visionary leadership. In 2008, the author and his students published the first Web-scale image annotation work in the proceedings of Advances in Multimedia Information Processing-PCM 2008, well before the research community widely adopted such directions with ImageNet. This book covers their groundbreaking work at Google Research between 2008 and 2011 on knowledge representation, semantic analysis of multimedia data, and scalability in signal extraction, data mining, and indexing.
The book is divided into two parts:
Part I - Data-Driven Knowledge Representation and Semantic Analysis: This section focuses on the key components of the mathematics of perception as it applies to data management and retrieval. Topics include data-driven feature selection/reduction, knowledge representation, semantic analysis, distance function formulation for measuring similarity, and multimodal fusion.
Part II - Machine Learning Scalability Issues: This section presents several parallel algorithms for scaling up machine learning for high-dimensional data and Web-scale datasets. It also includes real-world applications and provides insights into future research and development directions.
The general psychoevolutionary theory of emotion that is presented here has a number of important characteristics. First, it provides a broad evolutionary foundation for conceptualizing the domain of emotion as seen in animals and humans. Second, it provides a structural model which describes the interrelations among emotions. Third, it has demonstrated both theoretical and empirical relations among a number of derivative domains including personality traits, diagnoses, and ego defenses. Fourth, it has provided a theoretical rationale for the construction of tests and scales for the measurement of key dimensions within these various domains. Fifth, it has stimulated a good deal of empirical research using these tools and concepts. Finally, the theory provides useful insights into the relationships among emotions, adaptations, and evolution.
Defining “emotion” is a notorious problem. Without consensual conceptualization and operationalization of exactly what phenomenon is to be studied, progress in theory and research is difficult to achieve and fruitless debates are likely to proliferate. A particularly unfortunate example is William James’s asking the question “What is an emotion?” when he really meant “feeling”, a misnomer that started a debate which is still ongoing, more than a century later. This contribution attempts to sensitize researchers in the social and behavioral sciences to the importance of definitional issues and their consequences for distinguishing related but fundamentally different affective processes, states, and traits. Links between scientific and folk concepts of emotion are explored and ways to measure emotion and its components are discussed.
Entropy variation and information competence: Enhancing predictive accuracy of collaborative language models
- Y Edward
- Chang
Integrating emotional and linguistic models for ethical compliance in large language models
- Y Edward
- Chang
An explanation of in-context learning as implicit bayesian inference
- Michael Xie
Chateval: Towards better 11m-based evaluators through multi-agent debate
- Chi-Min Chan
- Weize Chen
- Yusheng Su
- Jianxuan Yu
- Wei Xue
- Shanghang Zhang
Towards guaranteed safe ai: A framework for ensuring robust and reliable ai systems
- David Dalrymple
- Joar Skalse
- Yoshua Bengio
- Stuart Russell
- Max Tegmark
- Sanjit Seshia
Basic Emotions, chapter 3
- Paul Ekman
Cooperation, competition, and maliciousness: Llm-stakeholders interactive negotiation
- Sahar Abdelnabi
- Amr Gomaa
- Sarath Sivaprasad
- Lea Schonherr
- Mario Fritz
Improving language model negotiation with self-play and in-context learning from ai feedback
- Yao Fu
- Hao Peng
- Tushar Khot
- Mirella Lapata
Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride
- June Price Tangney
- Kurt W Fischer
Encouraging divergent thinking in large language models through multi-agent debate
- Tian Liang
- Zhiwei He
- Wenxiang Jiao
- Xing Wang
- Yan Wang
- Rui Wang
Bowman. Debate helps supervise unreliable experts
- Julian Michael
- Salsabila Mahdi
- David Rein
- Jackson Petty
- Julien Dirani
- Vishakh Padmakumar
Should we be going mad? a look at multiagent debate strategies for llms
- Andries Smit
- Paul Duckworth
- Nathan Grinsztajn
- Barrett Thomas
- Arnu Pretorius