Search
Close this search box.

We are creating some awesome events for you. Kindly bear with us.

Logic-Aware Model Transforming Language Understanding

Image credits: news.mit.edu

Researchers have developed a logic-aware model that outperforms counterparts 500 times larger in specific language-understanding tasks without human-generated annotations. This model excels in performance while ensuring privacy and robustness, addressing concerns related to the inefficiency and privacy of large AI models.

Although Large Language Models (LLMs) have demonstrated promising abilities in generating language, art, and code, they come with high computational demands, and utilising application programming interfaces for data upload can pose risks to privacy. Smaller models have historically exhibited lesser capabilities, particularly in tasks involving multitasking and weak supervision, than their larger counterparts.

The researchers introduced the concept of “textual entailment” to aid in comprehending various language tasks by these models. In textual entailment, if one sentence (the premise) is true, then it is likely that the other sentence (the hypothesis) is also true. For instance, if the premise states “all cats have tails,” then the theory “a tabby cat has a tail” would be entailed by the premise.

The team’s previous research revealed that this approach, known as an “entailment model,” exhibited less bias than other language models. To leverage this concept, the researchers developed prompts that enable the models to determine if specific information is entailed by a given sentence or phrase across different tasks. This technique enhanced the model’s adaptability to diverse functions without requiring additional training, a phenomenon referred to as zero-shot adaptation.

In the domain of “natural language understanding,” numerous applications rely on discerning the relationship between two text pieces. For instance, in sentiment classification, the statement “I think the movie is good” can be inferred or entailed from a movie review stating, “I like the story and the acting is great,” indicating a positive sentiment. Similarly, in news classification, the topic of a news article can be inferred from its content. For example, the statement “the news article is about sports” can be entailed if the article’s main content reports on an NBA game. The researchers realised that many existing natural language understanding tasks could be reformulated as entailment tasks involving logical inference in natural language.

“Our research focuses on enhancing the capability of computer programs to comprehend and process natural language, which mimics the way humans speak and write,” explains Hongyin Luo, lead author of a new study from MIT CSAIL.

The study introduces entailment models with 350 million parameters that outperform supervised language models with 137 to 175 billion parameters without human-generated labels. This breakthrough can potentially revolutionise AI and machine learning, providing a scalable, reliable, and cost-effective solution for language modelling. Demonstrating the comparable performance of smaller models in language understanding opens avenues for sustainable and privacy-preserving AI technologies.

The model’s performance was enhanced through self-training, learning without human supervision or annotated data. This approach significantly improved results in sentiment analysis, question-answering, and news classification tasks. It surpassed Google’s LaMDA, FLAN, GPT models, and other supervised algorithms in zero-shot capabilities.

The research addresses the challenge of self-training in language models by developing a novel algorithm called ‘SimPLE’ (Simple Pseudo-Label Editing). By reviewing and modifying the initially generated pseudo-labels, the algorithm improves the overall quality of self-generated labels. CSAIL Senior Research Scientist James Glass emphasises that this study introduces an efficient approach for training large language models (LLMs) by framing language understanding tasks as contextual entailment problems and employing a self-training mechanism with pseudo-labelling. It enables the incorporation of substantial amounts of unlabeled text data during training.

“This study demonstrates the feasibility of developing relatively compact language models that excel in benchmark language understanding tasks when compared to models of similar or even larger sizes,” he concludes.

PARTNER

Qlik’s vision is a data-literate world, where everyone can use data and analytics to improve decision-making and solve their most challenging problems. A private company, Qlik offers real-time data integration and analytics solutions, powered by Qlik Cloud, to close the gaps between data, insights and action. By transforming data into Active Intelligence, businesses can drive better decisions, improve revenue and profitability, and optimize customer relationships. Qlik serves more than 38,000 active customers in over 100 countries.

PARTNER

As a Titanium Black Partner of Dell Technologies, CTC Global Singapore boasts unparalleled access to resources.

Established in 1972, we bring 52 years of experience to the table, solidifying our position as a leading IT solutions provider in Singapore. With over 300 qualified IT professionals, we are dedicated to delivering integrated solutions that empower your organization in key areas such as Automation & AI, Cyber Security, App Modernization & Data Analytics, Enterprise Cloud Infrastructure, Workplace Modernization and Professional Services.

Renowned for our consulting expertise and delivering expert IT solutions, CTC Global Singapore has become the preferred IT outsourcing partner for businesses across Singapore.

PARTNER

Planview has one mission: to build the future of connected work. Our solutions enable organizations to connect the business from ideas to impact, empowering companies to accelerate the achievement of what matters most. Planview’s full spectrum of Portfolio Management and Work Management solutions creates an organizational focus on the strategic outcomes that matter and empowers teams to deliver their best work, no matter how they work. The comprehensive Planview platform and enterprise success model enables customers to deliver innovative, competitive products, services, and customer experiences. Headquartered in Austin, Texas, with locations around the world, Planview has more than 1,300 employees supporting 4,500 customers and 2.6 million users worldwide. For more information, visit www.planview.com.

SUPPORTING ORGANISATION

SIRIM is a premier industrial research and technology organisation in Malaysia, wholly-owned by the Minister​ of Finance Incorporated. With over forty years of experience and expertise, SIRIM is mandated as the machinery for research and technology development, and the national champion of quality. SIRIM has always played a major role in the development of the country’s private sector. By tapping into our expertise and knowledge base, we focus on developing new technologies and improvements in the manufacturing, technology and services sectors. We nurture Small Medium Enterprises (SME) growth with solutions for technology penetration and upgrading, making it an ideal technology partner for SMEs.

PARTNER

HashiCorp provides infrastructure automation software for multi-cloud environments, enabling enterprises to unlock a common cloud operating model to provision, secure, connect, and run any application on any infrastructure. HashiCorp tools allow organizations to deliver applications faster by helping enterprises transition from manual processes and ITIL practices to self-service automation and DevOps practices. 

PARTNER

IBM is a leading global hybrid cloud and AI, and business services provider. We help clients in more than 175 countries capitalize on insights from their data, streamline business processes, reduce costs and gain the competitive edge in their industries. Nearly 3,000 government and corporate entities in critical infrastructure areas such as financial services, telecommunications and healthcare rely on IBM’s hybrid cloud platform and Red Hat OpenShift to affect their digital transformations quickly, efficiently and securely. IBM’s breakthrough innovations in AI, quantum computing, industry-specific cloud solutions and business services deliver open and flexible options to our clients. All of this is backed by IBM’s legendary commitment to trust, transparency, responsibility, inclusivity and service.