Contributors: Nicole Ren (GovTech), Ng Wei Cheng (GovTech)
VICA (Digital Clever Chat Assistant) is GovTech’s Digital Assistant platform that leverages Synthetic Intelligence (AI) to permit customers to create, practice and deploy chatbots on their web sites. On the time of writing, VICA helps over 100 chatbots and handles over 700,000 person queries in a month.
Behind the scenes, VICA’s NLP engine makes use of assorted applied sciences and frameworks starting from conventional intent-matching methods to generative AI frameworks like Retrieval Augmented Technology (RAG). By conserving updated with state-of-the-art applied sciences, our engine is consistently evolving, making certain that each citizen’s question will get matched to the absolute best reply.
Past easy Query-And-Reply (Q&A) capabilities, VICA goals to supercharge chatbots by way of conversational transactions. Our aim is to say goodbye to the robotic and awkward form-like expertise inside a chatbot, and say hey to personalised conversations with human-like help.
This text is the primary in a two half article sequence to share extra in regards to the generative AI options we’ve got inbuilt VICA. On this article, we’ll deal with how LLM brokers might help enhance the transaction course of in chatbots by way of utilizing LangChain’s Agent Framework.
- Introduction
- All about LangChain
- LangChain in production
- Challenges of productionizing LangChain
- Use case of LLM Agents
- Conclusion
- Find out more about VICA
- Acknowledgements
- References
Transaction-based chatbots are conversational brokers designed to facilitate and execute particular transactions for customers. These chatbots transcend easy Q&A interactions that happen by permitting customers to carry out duties similar to reserving, buying, or kind submission straight inside the chatbot interface.
With the intention to carry out transactions, the chatbots should be custom-made on the backend to deal with further person flows and make API calls.
With the rise of Massive Language Fashions (LLMs), it has opened new avenues for simplifying and enhancing the event of those options for chatbots. LLMs can vastly enhance a chatbot’s capability to understand and reply to a variety of queries, serving to to handle complicated transactions extra successfully.
Though intent-matching chatbot methods exist already to information customers by way of predefined flows for transactions, LLMs supply important benefits by sustaining context over multi-turn interactions and dealing with a variety of inputs and language variations. Beforehand, interactions typically felt awkward and stilted, as customers have been required to pick choices from premade playing cards or kind particular phrases as a way to set off a transaction circulate. For instance, a slight variation from “Can I make a cost?” to “Let me pay, please” might forestall the transaction circulate from triggering. In distinction, LLMs can adapt to numerous communication types permitting them to interpret person enter that doesn’t match neatly into predefined intents.
Recognizing this potential, our staff determined to leverage LLMs for transaction processing, enabling customers to enter transaction flows extra naturally and flexibly by breaking down and understanding their intentions. Provided that LangChain gives a framework for implementing agentic workflows, we selected to make the most of their agent framework to create an clever system to course of transactions.
On this article, we will even share two use circumstances we developed that make the most of LLM Brokers, specifically The Division of Statistics (DOS) Statistic Desk Builder, and the Pure Dialog Facility Reserving chatbot.
Earlier than we cowl how we made use of LLM Brokers to carry out transactions, we’ll first share on what’s LangChain and why we opted to experiment with this framework.
What’s LangChain?
LangChain is an open-source Python framework designed to help builders in constructing AI powered functions leveraging LLMs.
Why use LangChain?
The framework helps to simplify the event course of by offering abstractions and templates that allow fast software constructing, saving time and decreasing the necessity for our growth staff to code every thing from scratch. This permits for us to deal with higher-level performance and enterprise logic reasonably than low-level coding particulars. An instance of that is how LangChain helps to streamline third occasion integration with common service suppliers like MongoDB, OpenAI, and AWS, facilitating faster prototyping and decreasing the complexity of integrating varied companies. These abstractions not solely speed up growth but in addition enhance collaboration by offering a constant construction, permitting our staff to effectively construct, take a look at, and deploy AI functions.
What’s LangChain’s Agent Framework?
One of many most important options of utilizing Langchain is their agent framework. The framework permits for administration of clever brokers that work together with LLMs and different instruments to carry out complicated duties.
The three most important parts of the framework are
Brokers act as a reasoning engine as they resolve the suitable actions to take and the order to take these actions. They make use of an LLM to make the choices for them. An agent has an AgentExecutor that calls the agent and executes the instruments the agent chooses. It additionally takes the output of the motion and passes it to the agent till the ultimate final result is reached.
Instruments are interfaces that the agent could make use of. With the intention to create a instrument, a reputation and outline must be offered. The outline and identify of the instrument are vital as it is going to be added into the agent immediate. Which means the agent will resolve the instrument to make use of primarily based on the identify and outline offered.
A sequence consult with sequences of calls. The chain could be coded out steps or only a name to an LLM or a instrument. Chains could be custom-made or be used off-the-shelf primarily based on what LangChain supplies. A easy instance of a sequence is LLMChain, a sequence that run queries towards LLMs.
How did we use LangChain in VICA?
In VICA, we arrange a microservice for LangChain invoked by way of REST API. This helps to facilitate integration by permitting totally different parts of VICA to speak with LangChain independently. Consequently, we are able to effectively construct our LLM agent with out being affected by modifications or growth in different parts of the system.
LangChain as a framework is fairly in depth relating to the LLM area, protecting retrieval strategies, brokers and LLM analysis. Listed here are the parts we made use of when growing our LLM Agent.
ReAct Agent
In VICA, we made use of a single agent system. The agent makes use of ReAct logic to find out the sequence of actions to take (Yao et al., 2022). This immediate engineering approach will assist generate the next:
- Thought (Reasoning taken earlier than selecting the motion)
- Motion (Motion to take, typically a instrument)
- Motion Enter (Enter to the motion)
- Statement (Statement from the instrument output)
- Closing Reply (Generative closing reply that the agent returns)
> Getting into new AgentExecutor chain…
The person needs to know the climate in the present day
Motion: Climate Instrument
Motion Enter: "Climate in the present day"
Statement: Reply: "31 Levels Celsius, Sunny"
Thought: I now know the ultimate reply.
Closing Reply: The climate in the present day is sunny at 31 levels celsius.
> Completed chain.
Within the above instance, the agent was capable of perceive the person’s intention prior to picking the instrument to make use of. There was additionally verbal reasoning being generated that helps the mannequin plan the sequence of motion to take. If the commentary is inadequate to reply the query given, the agent can cycle to a special motion as a way to get nearer to the ultimate reply.
In VICA, we edited the agent immediate to higher go well with our use case. The bottom immediate offered by LangChain (link here) is usually adequate for commonest use circumstances, serving as an efficient start line. Nevertheless, it may be modified to reinforce efficiency and guarantee larger relevance to particular functions. This may be executed through the use of a customized immediate earlier than passing it as a parameter to the create_react_agent (is perhaps totally different primarily based in your model of LangChain).
To find out if our customized immediate was an enchancment, we employed an iterative immediate engineering method: Write, Consider and Refine (extra particulars right here). This course of ensured that the immediate generalized successfully throughout a broad vary of take a look at circumstances. Moreover, we used the bottom immediate offered by LangChain as a benchmark to judge our customized prompts, enabling us to evaluate their efficiency with various further context throughout varied transaction eventualities.
Customized Instruments & Chains (Immediate Chaining)
For the 2 customized chatbot options on this article, we made use of customized instruments that our Agent could make use of to carry out transactions. Our customized instruments make use of immediate chaining to breakdown and perceive a person’s request earlier than deciding what to do within the explicit instrument.
Immediate chaining is a method the place a number of prompts are utilized in sequence to deal with complicated duties or queries. It entails beginning with an preliminary immediate and utilizing its output as enter for subsequent prompts, permitting for iterative refinement and contextual continuity. This methodology enhances the dealing with of intricate queries, improves accuracy, and maintains coherence by progressively narrowing down the main target.
For every transaction use case, we broke the method into a number of steps, permitting us to present clearer directions to the LLM at every stage. This methodology improves accuracy by making duties extra particular and manageable. We can also inject localized context into the prompts, which clarifies the aims and enhances the LLM’s understanding. Based mostly on the LLM’s reasoning, our customized chains will make requests to exterior APIs to assemble knowledge to carry out the transaction.
At each step of immediate chaining, it’s essential to implement error dealing with, as LLMs can generally produce hallucinations or inaccurate responses. By incorporating error dealing with mechanisms similar to validation checks, we recognized and addressed inconsistencies or errors within the outputs. This allowed us to generate fallback responses to our customers that defined what the LLM did not purpose at.
Lastly, in our customized instrument, we kept away from merely utilizing the LLM generated output as the ultimate response because of the danger of hallucination. As a citizen going through chatbot, it’s essential to forestall our chatbots from disseminating any deceptive or inaccurate data. Subsequently, we be sure that all responses to person queries are derived from precise knowledge factors retrieved by way of our customized chains. We then format these knowledge factors into pre-defined responses, making certain that customers don’t see any direct output generated by the LLM.
Challenges of utilizing LLMs
Problem #1: Immediate chaining results in sluggish inference time
A problem with LLMs is their inference occasions. LLMs have excessive computational calls for because of their massive variety of parameters and having to be known as repeatedly for actual time processing, resulting in comparatively sluggish inference occasions (a couple of seconds per immediate). VICA is a chatbot that will get 700,000 queries in a month. To make sure person expertise, we purpose to offer our responses as shortly as attainable whereas making certain accuracy.
Immediate chaining will increase the consistency, controllability and reliability of LLM outputs. Nevertheless, every further chain we incorporate considerably slows down our resolution because it necessitates making an additional LLM request. To stability simplicity with effectivity, we set a tough restrict on the variety of chains to forestall extreme wait occasions for customers. We additionally opted to not use higher performing LLM fashions similar to GPT-4 because of their slower velocity, however opted for sooner however typically effectively performing LLMs.
Problem #2 :Hallucination
As seen within the latest incident with Google’s characteristic, AI Overview, having LLMs producing outputs can result in inaccurate or non-factual particulars. Though grounding the LLM makes it extra constant and fewer more likely to hallucinate, it doesn’t get rid of hallucination.
As talked about above, we made use of immediate chaining to carry out reasoning duties for transactions by breaking it down into smaller, simpler to grasp duties. By chaining LLMs, we’re capable of extract the knowledge wanted to course of complicated queries. Nevertheless, for the ultimate output, we crafted non-generative messages as the ultimate response from the reasoning duties that the LLM performs. Which means in VICA, our customers don’t see generated responses from our LLM Agent.
Problem #1: An excessive amount of abstraction
The primary situation with LangChain is that the framework abstracts away too many particulars, making it very troublesome to customise functions for particular actual world use circumstances.
With the intention to overcome such limitations, we needed to delve into the package deal and customise sure courses to higher go well with our use case. For example, we modified the AgentExecutor class to route the ReAct agent’s motion enter into the instrument that was chosen. This gave our customized instruments further context that helped with extracting data from person queries.
Problem #2: Lack of documentation
The second situation is the dearth of documentation and the continually evolving framework. This makes growth troublesome because it takes time to grasp how the framework works by way of wanting on the package deal code. There’s additionally a scarcity of consistency on how issues work, making it troublesome to select issues up as you go. Additionally with fixed updates on current courses, an improve in model can lead to beforehand working code out of the blue breaking.
If you’re planning to make use of LangChain in manufacturing, an recommendation could be to repair your manufacturing model and take a look at earlier than upgrading.
Use case #1: Division of Statistics (DOS) Desk builder
With regards to taking a look at statistical knowledge about Singapore, customers can discover it troublesome to seek out and analyze the knowledge that they’re on the lookout for. To deal with this situation, we got here up with a POC that goals to extract and current statistical knowledge in a desk format as a characteristic in our chatbot.
As DOS’s API is open for public use, we made use of the API documentation that was offered of their web site. Utilizing LLM’s pure language understanding capabilities, we handed the API documentation into the immediate. The LLM was then tasked to select the right API endpoint primarily based on what the statistical knowledge that the person was asking for. This meant that customers might ask for statistical data for annual/half-yearly/quarterly/month-to-month knowledge in proportion change/absolute values in a given time filter. For instance, we’re capable of question particular data similar to “GDP for Development in 2022” or “CPI in quarter 1 for the previous 3 years”.
We then did additional immediate chaining to interrupt the duty down much more, permitting for extra consistency in our closing output. The queries have been then processed to generate the statistics offered in a desk. As all the knowledge have been obtained from the API, not one of the numbers displayed are generated by LLMs thus avoiding any danger of spreading non-factual data.
Use case #2: Pure Dialog Facility Reserving Chatbot
In in the present day’s digital age, nearly all of bookings are performed by way of on-line web sites. Relying on the person interface, it might be a course of that entails sifting by way of quite a few dates to safe an obtainable slot, making it troublesome as you may have to look by way of a number of dates to seek out an obtainable reserving slot.
Reserving by way of pure dialog might simplify this course of. By simply typing one line similar to “I need to ebook a badminton courtroom at Fengshan at 9.30 am”, you’d have the ability to get a reserving or suggestions from a digital assistant.
With regards to reserving a facility, there are three issues we want from a person:
- The ability kind (e.g. Badminton, Assembly room, Soccer)
- Location (e.g. Ang Mo Kio, Maple Tree Enterprise Centre, Hive)
- Date (this week, 26 Feb, in the present day)
As soon as we’re capable of detect these data from pure language, we are able to create a customized reserving chatbot that’s reusable for a number of use circumstances (e.g. the reserving of hotdesk, reserving of sports activities amenities, and so forth).
The above instance illustrates a person inquiring in regards to the availability of a soccer discipline at 2.30pm. Nevertheless, the person is lacking a required data which is the date. Subsequently, the chatbot will ask a clarifying query to acquire the lacking date. As soon as the person supplies the date, the chatbot will course of this multi-turn dialog and try to seek out any obtainable reserving slots that matches the person’s request. As there was a reserving slot that matches the person’s precise description, the chatbot will current this data as a desk.
If there aren’t any obtainable reserving slots obtainable, our facility reserving chatbot would develop the search, exploring totally different timeslots or growing the search date vary. It could additionally try and advocate customers obtainable reserving slots primarily based on their earlier question if there their question leads to no obtainable bookings. This goals to reinforce the person expertise by eliminating the necessity to filter out unavailable dates when making a reserving, saving customers the trouble and time.
As a result of we use LLMs as our reasoning engine, a further profit is their multilingual capabilities, which allow them to purpose and reply to customers writing in several languages.
The instance above illustrates the chatbot’s capability to precisely course of the right facility, dates, and placement from the person’s message that was written in Korean to present the suitable non-generative response though there aren’t any obtainable slots for the date vary offered.
What we demonstrated was a quick instance of how our LLM Agent handles facility reserving transactions. In actuality, the precise resolution is much more complicated, having the ability to give a number of obtainable bookings for a number of areas, deal with postal codes, deal with areas too removed from the acknowledged location, and so forth. Though we wanted to make some modifications to the package deal to suit our particular use case, LangChain’s Agent Framework was helpful in serving to us chain a number of prompts collectively and use their outputs within the ReAct Agent.
Moreover, we designed this custom-made resolution to be simply extendable to any related reserving system that requires reserving by way of pure language.
On this first a part of our sequence, we explored how GovTech’s Digital Clever Chat Assistant (VICA) leverages LLM Brokers to reinforce chatbot capabilities, notably for transaction-based chatbots.
By integrating LangChain’s Agent Framework into VICA’s structure, we demonstrated its potential by way of the Division of Statistics (DOS) Desk Builder and Facility Reserving Chatbot use circumstances. These examples spotlight how LangChain can streamline complicated transaction interactions, enabling chatbots to deal with transaction associated duties like knowledge retrieval and reserving by way of pure dialog.
LangChain gives options to shortly develop and prototype subtle chatbot options, permitting builders to harness the ability of huge language fashions effectively. Nevertheless, challenges like inadequate documentation and extreme abstraction can result in elevated upkeep efforts as customizing the framework to suit particular wants could require important time and sources. Subsequently, evaluating an in-house resolution may supply larger long run customizability and stability.
Within the subsequent article, we can be protecting how chatbot engines could be improved by way of understanding multi-turn conversations.
Curious in regards to the potential of AI chatbots? If you’re a Singapore public service officer, you possibly can go to our web site at https://www.vica.gov.sg/ to create your personal customized chatbot and discover out extra!
Particular because of Wei Jie Kong for establishing necessities for the Facility Reserving Chatbot. We additionally want to thank Justin Wang and Samantha Yom, our hardworking interns, for his or her preliminary work on the DOS Desk builder.
Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, Ok., & Cao, Y. (2022). React: Synergizing reasoning and performing in language fashions. arXiv preprint arXiv:2210.03629.
