, authorized contracts are foundational paperwork that outline the relationships, obligations, and tasks between events. Whether or not it’s a partnership settlement, an NDA, or a provider contract, these paperwork typically include essential data that drives decision-making, threat administration, and compliance. Nevertheless, navigating and extracting insights from these contracts is usually a advanced and time-consuming course of.
On this submit, we’ll discover how we will streamline the method of understanding and dealing with authorized contracts by implementing an end-to-end resolution utilizing Agentic Graphrag. I see GraphRAG as an umbrella time period for any methodology that retrieves or causes over data saved in a data graph, enabling extra structured and context-aware responses.
By structuring authorized contracts right into a data graph in Neo4j, we will create a robust repository of knowledge that’s straightforward to question and analyze. From there, we’ll construct a LangGraph agent that permits customers to ask particular questions in regards to the contracts, making it potential to quickly uncover new insights.
The code is out there on this GitHub repository.
Why structuring information issues
Some domains work effectively with naive RAG, however authorized contracts current distinctive challenges.
As proven within the picture, relying solely on a vector index to retrieve related chunks can introduce dangers, corresponding to pulling data from irrelevant contracts. It is because authorized language is very structured, and comparable wording throughout completely different agreements can result in incorrect or deceptive retrieval. These limitations spotlight the necessity for a extra structured method, corresponding to GraphRAG, to make sure exact and context-aware retrieval.
To implement GraphRAG, we first have to assemble a data graph.
To construct a data graph for authorized contracts, we want a strategy to extract structured data from paperwork and retailer it alongside the uncooked textual content. An LLM may help by studying by means of contracts and figuring out key particulars corresponding to events, dates, contract sorts, and necessary clauses. As an alternative of treating the contract as only a block of textual content, we break it down into structured elements that mirror its underlying authorized which means. For instance, an LLM can acknowledge that “ACME Inc. agrees to pay $10,000 monthly beginning January 1, 2024” incorporates each a cost obligation and a begin date, which we will then retailer in a structured format.
As soon as we’ve got this structured information, we retailer it in a data graph, the place entities like firms, agreements, and clauses are represented as represented together with their relationships. The unstructured textual content stays accessible, however now we will use the structured layer to refine our searches and make retrieval way more exact. As an alternative of simply fetching essentially the most related textual content chunks, we will filter contracts primarily based on their attributes. This implies we will reply questions that naive RAG would wrestle with, corresponding to what number of contracts have been signed final month or whether or not we’ve got any energetic agreements with a particular firm. These questions require aggregation and filtering, which isn’t potential with normal vector-based retrieval alone.
By combining structured and unstructured information, we additionally make retrieval extra context-aware. If a consumer asks a couple of contract’s cost phrases, we be sure that the search is constrained to the proper settlement relatively than counting on textual content similarity, which could pull in phrases from unrelated contracts. This hybrid method overcomes the restrictions of naive RAG and permits for a a lot deeper and extra dependable evaluation of authorized paperwork.
Graph building
We’ll leverage an LLM to extract structured data from authorized paperwork, utilizing the CUAD (Contract Understanding Atticus Dataset), a broadly used benchmark dataset for contract evaluation licensed underneath CC BY 4.0. CUAD dataset incorporates over 500 contracts, making it a great dataset for evaluating our structured extraction pipeline.
The token depend distribution for the contracts is visualized under.
Most contracts on this dataset are comparatively brief, with token counts under 10,000. Nevertheless, there are some for much longer contracts, with a number of reaching as much as 80,000 tokens. These lengthy contracts are uncommon, whereas shorter ones make up the bulk. The distribution reveals a steep drop-off, which means lengthy contracts are the exception relatively than the rule.
We’re utilizing Gemini-2.0-Flash for extraction, which has a 1 million token enter restrict, so dealing with these contracts isn’t an issue. Even the longest contracts in our dataset (round 80,000 tokens) match effectively throughout the mannequin’s capability. Since most contracts are a lot shorter, we don’t have to fret about truncation or breaking paperwork into smaller chunks for processing.
Structured information extraction
Most industrial LLMs have the choice to make use of Pydantic objects to outline the schema of the output. An instance for location:
class Location(BaseModel):
"""
Represents a bodily location together with handle, metropolis, state, and nation.
"""
handle: Elective[str] = Subject(
..., description="The road handle of the situation.Use None if not offered"
)
metropolis: Elective[str] = Subject(
..., description="The town of the situation.Use None if not offered"
)
state: Elective[str] = Subject(
..., description="The state or area of the situation.Use None if not offered"
)
nation: str = Subject(
...,
description="The nation of the situation. Use the two-letter ISO normal.",
)When utilizing LLMs for structured output, Pydantic helps outline a transparent schema by specifying the kinds of attributes and offering descriptions that information the mannequin’s responses. Every discipline has a sort, corresponding to str or Elective[str], and an outline that tells the LLM precisely methods to format the output.
For instance, in a Location mannequin, we outline key attributes like handle, metropolis, state, and nation, specifying what information is predicted and the way it must be structured. The nation discipline, for example, follows two-letter nation code normal like "US", "FR", or "JP", as an alternative of inconsistent variations like “United States” or “USA.” This precept applies to different structured information as effectively, ISO 8601 retains dates in an ordinary format (YYYY-MM-DD), and so forth.
By defining structured output with Pydantic, we make LLM responses extra dependable, machine-readable, and simpler to combine into databases or APIs. Clear discipline descriptions additional assist the mannequin generate accurately formatted information, decreasing the necessity for post-processing.
The Pydantic schema fashions may be extra refined just like the Contract mannequin under, which captures key particulars of a authorized settlement, guaranteeing the extracted information follows a standardized construction.
class Contract(BaseModel):
"""
Represents the important thing particulars of the contract.
"""
abstract: str = Subject(
...,
description=("Excessive degree abstract of the contract with related details and particulars. Embody all related data to offer full image."
"Do no use any pronouns"),
)
contract_type: str = Subject(
...,
description="The kind of contract being entered into.",
enum=CONTRACT_TYPES,
)
events: Listing[Organization] = Subject(
...,
description="Listing of events concerned within the contract, with particulars of every social gathering's position.",
)
effective_date: str = Subject(
...,
description=(
"Enter the date when the contract turns into efficient in yyyy-MM-dd format."
"If solely the yr (e.g., 2015) is thought, use 2015-01-01 because the default date."
"All the time fill in full date"
),
)
contract_scope: str = Subject(
...,
description="Description of the scope of the contract, together with rights, duties, and any limitations.",
)
period: Elective[str] = Subject(
None,
description=(
"The period of the settlement, together with provisions for renewal or termination."
"Use ISO 8601 durations normal"
),
)
end_date: Elective[str] = Subject(
None,
description=(
"The date when the contract expires. Use yyyy-MM-dd format."
"If solely the yr (e.g., 2015) is thought, use 2015-01-01 because the default date."
"All the time fill in full date"
),
)
total_amount: Elective[float] = Subject(
None, description="Whole worth of the contract."
)
governing_law: Elective[Location] = Subject(
None, description="The jurisdiction's legal guidelines governing the contract."
)
clauses: Elective[List[Clause]] = Subject(
None, description=f"""Related summaries of clause sorts. Allowed clause sorts are {CLAUSE_TYPES}"""
)This contract schema organizes key particulars of authorized agreements in a structured method, making it simpler to investigate with LLMs. It contains various kinds of clauses, corresponding to confidentiality or termination, every with a brief abstract. The events concerned are listed with their names, areas, and roles, whereas contract particulars cowl issues like begin and finish dates, whole worth, and governing legislation. Some attributes, corresponding to governing legislation, may be outlined utilizing nested fashions, enabling extra detailed and sophisticated outputs.
The nested object method works effectively with some AI fashions that deal with advanced information relationships, whereas others might wrestle with deeply nested particulars.
We will take a look at our method utilizing the next instance. We’re utilizing the LangChain framework to orchestrate LLMs.
llm = ChatGoogleGenerativeAI(mannequin="gemini-2.0-flash")
llm.with_structured_output(Contract).invoke(
"Tomaz works with Neo4j since 2017 and can make a billion greenback till 2030."
"The contract was signed in Las Vegas"
)which outputs
Contract(
abstract="Tomaz works with Neo4j since 2017 and can make a billion greenback till 2030.",
contract_type="Service",
events=[
Organization(
name="Tomaz",
location=Location(
address=None,
city="Las Vegas",
state=None,
country="US"
),
role="employee"
),
Organization(
name="Neo4j",
location=Location(
address=None,
city=None,
state=None,
country="US"
),
role="employer"
)
],
effective_date="2017-01-01",
contract_scope="Tomaz will work with Neo4j",
period=None,
end_date="2030-01-01",
total_amount=1_000_000_000.0,
governing_law=None,
clauses=None
)Now that our contract information is in a structured format, we will outline the Cypher question wanted to import it into Neo4j, mapping entities, relationships, and key clauses right into a graph construction. This step transforms uncooked extracted information right into a queryable data graph, enabling environment friendly traversal and retrieval of contract insights.
UNWIND $information AS row
MERGE (c:Contract {file_id: row.file_id})
SET c.abstract = row.abstract,
c.contract_type = row.contract_type,
c.effective_date = date(row.effective_date),
c.contract_scope = row.contract_scope,
c.period = row.period,
c.end_date = CASE WHEN row.end_date IS NOT NULL THEN date(row.end_date) ELSE NULL END,
c.total_amount = row.total_amount
WITH c, row
CALL (c, row) {
WITH c, row
WHERE row.governing_law IS NOT NULL
MERGE (c)-[:HAS_GOVERNING_LAW]->(l:Location)
SET l += row.governing_law
}
FOREACH (social gathering IN row.events |
MERGE (p:Occasion {title: social gathering.title})
MERGE (p)-[:HAS_LOCATION]->(pl:Location)
SET pl += social gathering.location
MERGE (p)-[pr:PARTY_TO]->(c)
SET pr.position = social gathering.position
)
FOREACH (clause IN row.clauses |
MERGE (c)-[:HAS_CLAUSE]->(cl:Clause {kind: clause.clause_type})
SET cl.abstract = clause.abstract
)This Cypher question imports structured contract information into Neo4j by creating Contract nodes with attributes corresponding to abstract, contract_type, effective_date, period, and total_amount. If a governing legislation is specified, it hyperlinks the contract to a Location node. Events concerned within the contract are saved as Occasion nodes, with every social gathering related to a Location and assigned a task in relation to the contract. The question additionally processes clauses, creating Clause nodes and linking them to the contract whereas storing their kind and abstract.
After processing and importing the contracts, the ensuing graph follows the next graph schema.
Let’s additionally check out a single contract.
This graph represents a contract construction the place a contract (orange node) connects to numerous clauses (crimson nodes), events (blue nodes), and areas (violet nodes). The contract has three clauses: Renewal & Termination, Legal responsibility & Indemnification, and Confidentiality & Non-Disclosure. Two events, Modus Media Worldwide and Dragon Methods, Inc., are concerned, every linked to their respective areas, Netherlands (NL) and United States (US). The contract is ruled by U.S. legislation. The contract node additionally incorporates extra metadata, together with dates and different related particulars.
A public read-only occasion containing CUAD authorized contracts is out there with the next credentials.
URI: neo4j+s://demo.neo4jlabs.com
username: legalcontracts
password: legalcontracts
database: legalcontractsEntity decision
Entity decision in authorized contracts is difficult resulting from variations in how firms, people, and areas are referenced. An organization would possibly seem as “Acme Inc.” in a single contract and “Acme Company” in one other, requiring a course of to find out whether or not they seek advice from the identical entity.
One method is to generate candidate matches utilizing textual content embeddings or string distance metrics like Levenshtein distance. Embeddings seize semantic similarity, whereas string distance measures character-level variations. As soon as candidates are recognized, extra analysis is required, evaluating metadata corresponding to addresses or tax IDs, analyzing shared relationships within the graph, or incorporating human evaluate for essential circumstances.
For resolving entities at scale, each open-source options like Dedupe and industrial instruments like Senzing supply automated strategies. Selecting the best method relies on information high quality, accuracy necessities, and whether or not guide oversight is possible.
With the authorized graph constructed, we will transfer onto the agentic GraphRAG implementation.
Agentic GraphRAG
Agentic architectures range broadly in complexity, modularity, and reasoning capabilities. At their core, these architectures contain an LLM performing as a central reasoning engine, typically supplemented with instruments, reminiscence, and orchestration mechanisms. The important thing differentiator is how a lot autonomy the LLM has in making selections and the way interactions with exterior programs are structured.
One of many easiest and only designs, significantly for chatbot-like implementations, is a direct LLM-with-tools method. On this setup, the LLM serves because the decision-maker, dynamically choosing which instruments to invoke (if any), retrying operations when needed, and executing a number of instruments in sequence to meet advanced requests.
The diagram represents a easy LangGraph agent workflow. It begins at __start__, transferring to the assistant node, the place the LLM processes consumer enter. From there, the assistant can both name instruments to fetch related data or transition on to __end__ to finish the interplay. If a software is used, the assistant processes the response earlier than deciding whether or not to name one other software or finish the session. This construction permits the agent to autonomously decide when exterior data is required earlier than responding.
This method is especially well-suited to stronger industrial fashions like Gemini or GPT-4o, which excel at reasoning and self-correction.
Instruments
LLMs are highly effective reasoning engines, however their effectiveness typically relies on how effectively they’re outfitted with exterior instruments. These instruments , whether or not database queries, APIs, or search capabilities, lengthen an LLM’s skill to retrieve details, carry out calculations, or work together with structured information.
Designing instruments which can be each normal sufficient to deal with numerous queries and exact sufficient to return significant outcomes is extra artwork than science. What we’re actually constructing is a semantic layer between the LLM and the underlying information. Reasonably than requiring the LLM to know the precise construction of a Neo4j data graph or a database schema, we outline instruments that summary away these complexities.
With this method, the LLM doesn’t have to know whether or not contract data is saved as graph nodes and relationships or as uncooked textual content in a doc retailer. It solely must invoke the proper software to fetch related information primarily based on a consumer’s query.
In our case, the contract retrieval software serves as this semantic interface. When a consumer asks about contract phrases, obligations, or events, the LLM calls a structured question software that interprets the request right into a database question, retrieves related data, and presents it in a format the LLM can interpret and summarize. This allows a versatile, model-agnostic system the place completely different LLMs can work together with contract information with no need direct data of its storage or construction.
There’s no one-size-fits-all normal for designing an optimum toolset. What works effectively for one mannequin might fail for an additional. Some fashions deal with ambiguous software directions gracefully, whereas others wrestle with advanced parameters or require specific prompting. The trade-off between generality and task-specific effectivity means software design requires iteration, testing, and fine-tuning for the LLM in use.
For contract evaluation, an efficient software ought to retrieve contracts and summarize key phrases with out requiring customers to phrase queries rigidly. Reaching this flexibility relies on considerate immediate engineering, sturdy schema design, and adaptation to completely different LLM capabilities. As fashions evolve, so do methods for making instruments extra intuitive and efficient.
On this part, we’ll discover completely different approaches to software implementation, evaluating their flexibility, effectiveness, and compatibility with varied LLMs.
My most well-liked method is to dynamically and deterministically assemble a Cypher question and execute it towards the database. This methodology ensures constant and predictable question era whereas sustaining implementation flexibility. By structuring queries this fashion, we reinforce the semantic layer, permitting consumer inputs to be seamlessly translated into database retrievals. This retains the LLM targeted on retrieving related data relatively than understanding the underlying information mannequin.
Our software is meant to establish related contracts, so we have to present the LLM with choices to look contracts primarily based on varied attributes. The enter description is once more offered as a Pydantic object.
class ContractInput(BaseModel):
min_effective_date: Elective[str] = Subject(
None, description="Earliest contract efficient date (YYYY-MM-DD)"
)
max_effective_date: Elective[str] = Subject(
None, description="Newest contract efficient date (YYYY-MM-DD)"
)
min_end_date: Elective[str] = Subject(
None, description="Earliest contract finish date (YYYY-MM-DD)"
)
max_end_date: Elective[str] = Subject(
None, description="Newest contract finish date (YYYY-MM-DD)"
)
contract_type: Elective[str] = Subject(
None, description=f"Contract kind; legitimate sorts: {CONTRACT_TYPES}"
)
events: Elective[List[str]] = Subject(
None, description="Listing of events concerned within the contract"
)
summary_search: Elective[str] = Subject(
None, description="Examine abstract of the contract"
)
nation: Elective[str] = Subject(
None, description="Nation the place the contract applies. Use the two-letter ISO normal."
)
energetic: Elective[bool] = Subject(None, description="Whether or not the contract is energetic")
monetary_value: Elective[MonetaryValue] = Subject(
None, description="The entire quantity or worth of a contract"
)With LLM instruments, attributes can take varied types relying on their objective. Some fields are easy strings, corresponding to contract_type and nation, which retailer single values. Others, like events, are lists of strings, permitting a number of entries (e.g., a number of entities concerned in a contract).
Past primary information sorts, attributes also can signify advanced objects. For instance, monetary_value makes use of a MonetaryValue object, which incorporates structured information corresponding to forex kind and the operator. Whereas attributes with nested objects supply a transparent and structured illustration of information, fashions are inclined to wrestle to deal with them successfully, so we should always maintain them easy.
As a part of this undertaking, we’re experimenting with a further cypher_aggregation attribute, offering the LLM with better flexibility for eventualities that require particular filtering or aggregation.
cypher_aggregation: Elective[str] = Subject(
None,
description="""Customized Cypher assertion for superior aggregations and analytics.
This will likely be appended to the bottom question:
```
MATCH (c:Contract)
WITH c, abstract, contract_type, contract_scope, effective_date, end_date, events, energetic, monetary_value, contract_id, nations
```
Examples:
1. Depend contracts by kind:
```
RETURN contract_type, depend(*) AS depend ORDER BY depend DESC
```
2. Calculate common contract period by kind:
```
WITH contract_type, effective_date, end_date
WHERE effective_date IS NOT NULL AND end_date IS NOT NULL
WITH contract_type, period.between(effective_date, end_date).days AS period
RETURN contract_type, avg(period) AS avg_duration ORDER BY avg_duration DESC
```
3. Calculate contracts per efficient date yr:
```
RETURN effective_date.yr AS yr, depend(*) AS depend ORDER BY yr
```
4. Counts the social gathering with the best variety of energetic contracts:
```
UNWIND events AS social gathering
WITH social gathering.title AS party_name, energetic, depend(*) AS contract_count
WHERE energetic = true
RETURN party_name, contract_count
ORDER BY contract_count DESC
LIMIT 1
```
""" The cypher_aggregation attribute permits LLMs to outline customized Cypher statements for superior aggregations and analytics. It extends the bottom question by appending question-specified aggregation logic, enabling versatile filtering and computation.
This function helps use circumstances corresponding to counting contracts by kind, calculating common contract period, analyzing contract distributions over time, and figuring out key events primarily based on contract exercise. By leveraging this attribute, the LLM can dynamically generate insights tailor-made to particular analytical wants with out requiring predefined question constructions.
Whereas this flexibility is effective, it must be fastidiously evaluated, as elevated adaptability comes at the price of diminished consistency and robustness because of the added complexity of the operation.
We should clearly outline the perform’s title and outline when presenting it to the LLM. A well-structured description helps information the mannequin in utilizing the perform accurately, guaranteeing it understands its objective, anticipated inputs, and outputs. This reduces ambiguity and improves the LLM’s skill to generate significant and dependable queries.
class ContractSearchTool(BaseTool):
title: str = "ContractSearch"
description: str = (
"helpful for when you could reply questions associated to any contracts"
)
args_schema: Kind[BaseModel] = ContractInputLastly, we have to implement a perform that processes the given inputs, constructs the corresponding Cypher assertion, and executes it effectively.
The core logic of the perform facilities on developing the Cypher assertion. We start by matching the contract as the muse of the question.
cypher_statement = "MATCH (c:Contract) "Subsequent, we have to implement the perform that processes the enter parameters. On this instance, we primarily use attributes to filter contracts primarily based on the given standards.
Easy property filtering
For instance, the contract_type attribute is used to carry out easy node property filtering.
if contract_type:
filters.append("c.contract_type = $contract_type")
params["contract_type"] = contract_typeThis code provides a Cypher filter for contract_type whereas utilizing question parameters for values to forestall question injection safety subject.
Because the potential contract kind values are introduced within the attribute description
contract_type: Elective[str] = Subject(
None, description=f"Contract kind; legitimate sorts: {CONTRACT_TYPES}"
)we don’t have to fret about mapping values from enter to legitimate contract sorts because the LLM will deal with that.
Inferred property filtering
We’re constructing instruments for an LLM to work together with a data graph, the place the instruments function an abstraction layer over structured queries. A key function is the power to make use of inferred properties at runtime, just like an ontology however dynamically computed.
if energetic is just not None:
operator = ">=" if energetic else "<"
filters.append(f"c.end_date {operator} date()")Right here, energetic acts as a runtime classification, figuring out whether or not a contract is ongoing (>= date()) or expired (< date()). This logic extends structured KG queries by computing properties solely when wanted, enabling extra versatile LLM reasoning. By dealing with logic like this inside instruments, we make sure the LLM interacts with simplified, intuitive operations, preserving it targeted on reasoning relatively than question formulation.
Neighbor filtering
Typically filtering relies on neighboring nodes, corresponding to limiting outcomes to contracts involving particular events. The events attribute is an non-obligatory listing, and when offered, it ensures solely contracts linked to these entities are thought-about:
if events:
parties_filter = []
for i, social gathering in enumerate(events):
party_param_name = f"party_{i}"
parties_filter.append(
f"""EXISTS {{
MATCH (c)<-[:PARTY_TO]-(social gathering)
WHERE toLower(social gathering.title) CONTAINS ${party_param_name}
}}"""
)
params[party_param_name] = social gathering.decrease()This code filters contracts primarily based on their related events, treating the logic as AND, which means all specified circumstances should be met for a contract to be included. It iterates by means of the offered events listing and constructs a question the place every social gathering situation should maintain.
For every social gathering, a novel parameter title is generated to keep away from conflicts. The EXISTS clause ensures that the contract has a PARTY_TO relationship to a celebration whose title incorporates the required worth. The title is transformed to lowercase to permit case-insensitive matching. Every social gathering situation is added individually, implementing an implicit AND between them.
If extra advanced logic have been wanted, corresponding to supporting OR circumstances or permitting completely different matching standards, the enter would wish to alter. As an alternative of a easy listing of social gathering names, a structured enter format specifying operators can be required.
Moreover, we might implement a party-matching methodology that tolerates minor typos, bettering the consumer expertise by dealing with variations in spelling and formatting.
Customized operator filtering
So as to add extra flexibility, we will introduce an operator object as a nested attribute, permitting extra management over filtering logic. As an alternative of hardcoding comparisons, we outline an enumeration for operators and use it dynamically.
For instance, with financial values, a contract would possibly must be filtered primarily based on whether or not its whole quantity is bigger than, lower than, or precisely equal to a specified worth. As an alternative of assuming a set comparability logic, we outline an enum that represents the potential operators:
class NumberOperator(str, Enum):
EQUALS = "="
GREATER_THAN = ">"
LESS_THAN = "<"
class MonetaryValue(BaseModel):
"""The entire quantity or worth of a contract"""
worth: float
operator: NumberOperator
if monetary_value:
filters.append(f"c.total_amount {monetary_value.operator.worth} $total_value")
params["total_value"] = monetary_value.worthThis method makes the system extra expressive. As an alternative of inflexible filtering guidelines, the software interface permits the LLM to specify not only a worth however the way it must be in contrast, making it simpler to deal with a broader vary of queries whereas preserving the LLM’s interplay easy and declarative.
Some LLMs wrestle with nested objects as inputs, making it tougher to deal with structured operator-based filtering. Including a between operator introduces extra complexity because it requires two separate values, which may result in ambiguity in parsing and enter validation.
Min and Max attributes
To maintain issues easier, I are inclined to gravitate towards utilizing min and max attributes for dates, as this naturally helps vary filtering and makes the between logic easy.
if min_effective_date:
filters.append("c.effective_date >= date($min_effective_date)")
params["min_effective_date"] = min_effective_date
if max_effective_date:
filters.append("c.effective_date <= date($max_effective_date)")
params["max_effective_date"] = max_effective_dateThis perform filters contracts primarily based on an efficient date vary by including an non-obligatory decrease and higher certain situation when min_effective_date and max_effective_date are offered, guaranteeing that solely contracts throughout the specified date vary are included.
Semantic search
An attribute will also be used for semantic search, the place as an alternative of counting on a vector index upfront, we use a post-filtering method to metadata filtering. First, structured filters, like date ranges, financial values, or events, are utilized to slender down the candidate set. Then, vector search is carried out over this filtered subset to rank outcomes primarily based on semantic similarity.
if summary_search:
cypher_statement += (
"WITH c, vector.similarity.cosine(c.embedding, $embedding) "
"AS rating ORDER BY rating DESC WITH c, rating WHERE rating > 0.9 "
) # Outline a threshold restrict
params["embedding"] = embeddings.embed_query(summary_search)
else: # Else we kind by newest
cypher_statement += "WITH c ORDER BY c.effective_date DESC "This code applies semantic search when summary_search is offered by computing cosine similarity between the contract’s embedding and the question embedding, ordering outcomes by relevance, and filtering out low-scoring matches with a threshold of 0.9. In any other case, it defaults to sorting contracts by the latest effective_date.
Dynamic queries
The cypher aggregation attribute is an experiment I needed to check that offers the LLM a level of partial text2cypher functionality, permitting it to dynamically generate aggregations after the preliminary structured filtering. As an alternative of predefining each potential aggregation, this method lets the LLM specify calculations like counts, averages, or grouped summaries on demand, making queries extra versatile and expressive. Nevertheless, since this shifts extra question logic to the LLM, guaranteeing all generated queries work accurately turns into difficult, as malformed or incompatible Cypher statements can break execution. This trade-off between flexibility and reliability is a key consideration in designing the system.
if cypher_aggregation:
cypher_statement += """WITH c, c.abstract AS abstract, c.contract_type AS contract_type,
c.contract_scope AS contract_scope, c.effective_date AS effective_date, c.end_date AS end_date,
[(c)<-[r:PARTY_TO]-(social gathering) | {social gathering: social gathering.title, position: r.position}] AS events, c.end_date >= date() AS energetic, c.total_amount as monetary_value, c.file_id AS contract_id,
apoc.coll.toSet([(c)<-[:PARTY_TO]-(social gathering)-[:LOCATED_IN]->(nation) | nation.title]) AS nations """
cypher_statement += cypher_aggregationIf no cypher aggregation is offered, we return the overall depend of recognized contracts together with solely 5 instance contracts to keep away from overwhelming the immediate. Dealing with extreme rows is essential, as an LLM battling a large outcome set isn’t helpful. Moreover, LLM producing solutions with 100 contract titles isn’t an excellent consumer expertise both.
cypher_statement += """WITH gather(c) AS nodes
RETURN {
total_count_of_contracts: measurement(nodes),
example_values: [
el in nodes[..5] |
{abstract:el.abstract, contract_type:el.contract_type,
contract_scope: el.contract_scope, file_id: el.file_id,
effective_date: el.effective_date, end_date: el.end_date,
monetary_value: el.total_amount, contract_id: el.file_id,
events: [(el)<-[r:PARTY_TO]-(social gathering) | {title: social gathering.title, position: r.position}],
nations: apoc.coll.toSet([(el)<-[:PARTY_TO]-()-[:LOCATED_IN]->(nation) | nation.title])}
]
} AS output"""This cypher assertion collects all matching contracts into an inventory, returning the overall depend and as much as 5 instance contracts with key attributes, together with abstract, kind, scope, dates, financial worth, related events with roles, and distinctive nation areas.
Now that our contract search software is constructed, we hand it off to the LLM and similar to that, we’ve got agentic GraphRAG carried out.
Agent Benchmark
For those who’re severe about implementing agentic GraphRAG, you want an analysis dataset, not simply as a benchmark however as a basis for your entire undertaking. A well-constructed dataset helps outline the scope of what the system ought to deal with, guaranteeing that preliminary growth aligns with real-world use circumstances. Past that, it turns into a useful software for evaluating efficiency, permitting you to measure how effectively the LLM interacts with the graph, retrieves data, and applies reasoning. It’s additionally important for immediate engineering optimizations, letting you iteratively refine queries, software use, and response formatting with clear suggestions relatively than guesswork. And not using a structured dataset, you’re flying blind, making enhancements tougher to quantify and inconsistencies harder to catch.
The code for the benchmark is accessible on GitHub.
I’ve compiled an inventory of twenty-two questions which we’ll use to judge the system. Moreover, we’re going to introduce a brand new metric referred to as answer_satisfaction the place we will likely be present a customized immediate.
answer_satisfaction = AspectCritic(
title="answer_satisfaction",
definition="""You'll consider an ANSWER to a authorized QUESTION primarily based on a offered SOLUTION.
Price the reply on a scale from 0 to 1, the place:
- 0 = incorrect, considerably incomplete, or deceptive
- 1 = right and sufficiently full
Take into account these analysis standards:
1. Factual correctness is paramount - the reply should not contradict the answer
2. The reply should handle the core components of the answer
3. Extra related data past the answer is suitable and will improve the reply
4. Technical authorized terminology must be used appropriately if current within the resolution
5. For quantitative authorized analyses, correct figures should be offered
+ fewshots
"""Many questions can return a considerable amount of data. For instance, asking for contracts signed earlier than 2020 would possibly yield tons of of outcomes. Because the LLM receives each the overall depend and some instance entries, our analysis ought to concentrate on the overall depend, relatively than which particular examples the LLM chooses to indicate.
The offered outcomes point out that every one evaluated fashions (Gemini 1.5 Professional, Gemini 2.0 Flash, and GPT-4o) carry out equally effectively for many software calls, with GPT-4o barely outperforming the Gemini fashions (0.82 vs. 0.77). The noticeable distinction emerges primarily when partial text2cypher is used, significantly for varied aggregation operations.
Word that that is solely 22 pretty easy questions, so we didn’t actually discover reasoning capabilities of LLMs.
Moreover, I’ve seen tasks the place accuracy may be improved considerably by leveraging Python for aggregations, as LLMs usually deal with Python code era and execution higher than producing advanced Cypher queries straight.
Internet Software
We’ve additionally constructed a easy React net utility, powered by LangGraph hosted on FastAPI, which streams responses on to the frontend. Particular because of Anej Gorkic for creating the online app.
You’ll be able to launch your entire stack with the next command:
docker compose upAnd navigate to localhost:5173
Abstract
As LLMs acquire stronger reasoning capabilities, they, when paired with the proper instruments, can develop into highly effective brokers for navigating advanced domains like authorized contracts. On this submit, we’ve solely scratched the floor, specializing in core contract attributes whereas barely touching the wealthy number of clauses present in real-world agreements. There’s vital room for development, from increasing clause protection to refining software design and interplay methods.
The code is out there on GitHub.
Photos
All pictures on this submit have been created by the creator.
