What are we doing, and what are we assuming while we do it?
v2 extends LLM Ambiguity Lab beyond task resolution. It first resolves what action is being requested, then maps the facts, interpretations, concerns, and assumptions carried into that action before execution.
Different framings may remain explicitly preserved.
A factual claim can be mapped without being independently verified.
1. Two-Layer Task & Premise Simulator
Enter an instruction or choose a preset. Layer 1 resolves the task. Layer 2 maps premise structure only when the task is sufficiently resolved.
The task can be represented without adding premise structure beyond the operational request.
2. Fact / View / Care
v2 classifies the role a premise plays in the task. Classification is not a verdict on truth, reasonableness, or moral correctness.
A claim about what is, was, or happened
A proposition functioning as a factual claim in the task.
FACT means “factual claim,” not “verified true.”
An interpretation or framing
An evaluation, causal explanation, judgment, or perspective that shapes how the task is understood.
“The policy failed.” · “Leadership ignored the warnings.”
What matters within the task
A concern, protected interest, priority, value, or desired outcome.
CARE is not treated as a factual proposition requiring proof.
3. Premise Mapping Fields
Each premise is represented with enough structure to show where it came from, how strongly it affects execution, and whether the system is adding interpretation of its own.
source
Who or what introduced the premise.
user_implied
provided_source
system_default
agent_inference
unknown
materiality
Whether changing the premise would materially change execution.
support_state
How the proposition is currently represented, not a universal truth score.
unsupported
disputed
unknown
not_applicable
4. Premise Alignment Policy
The premise layer intentionally avoids a pseudo-probability score. The question is whether the premise structure is sufficient for the next action.
Represented well enough to proceed
Relevant premises are sufficiently identified and no unresolved material premise prevents execution.
Different framings remain
Preserve the differences explicitly. Agreement is not required when the task can continue without resolving them.
The treatment of a premise changes the output
Ask a focused question that distinguishes the materially different interpretations.
A necessary premise is missing
Expose the missing premise rather than silently inventing it.
5. Divergence & Clarification
v2 distinguishes “we understand the difference” from “we agree.” The former is often enough.
| Anti-pattern | Failure | Preferred behavior |
|---|---|---|
| FACT = truth | A user's assertion is silently upgraded into verified reality. | Separate functional role from verification. |
| Agent inference = user statement | The system's interpretation is laundered into user intent. | Expose premise source. |
| Alignment = agreement | Different positions are compressed into synthetic consensus. | Preserve divergence when possible. |
| Clarify everything | Minor uncertainty becomes conversational friction. | Interrupt only for material unresolved premises. |
| Silent narrative replacement | The system rewrites the task around its own assumptions. | Disclose assumptions and preserve attribution. |
6. Premise-Aware Execution Handoff
The public simulator does not execute the downstream task. It shows the structured object an execution layer could receive.
7. How This v2 Demo Works
Deterministic presets
The six reference examples return predefined, inspectable results. They are the canonical teaching cases.
Limited free-text heuristics
Free text uses documented lexical and structural signals. It is not a general semantic parser or a model-confidence detector.
No truth engine
v2 maps premise roles and support states. It does not independently establish objective truth.
8. What This Lab Does Not Do
- It does not infer hidden motives or diagnose people.
- It does not reveal hidden model reasoning.
- It does not determine objective truth automatically.
- It does not decide which participant is morally correct.
- It does not force disagreement into consensus.
- It does not guarantee safe downstream execution.
- It does not replace legal, medical, financial, or other domain expertise.
- It does not turn missing premises into facts.
How v2 Extends v1
v1 asks: “What are we doing?”
v2 adds: “What are we assuming while we do it?”
DCRL remains the upstream operational-ambiguity layer. SANA premise alignment maps what remains after the task is resolved. The two layers are deliberately separable.