[https://nvbugs/6550126][fix] Renamed the integration test's kwarg and its constant to…

[trtllm-agent]

Summary

• Root cause: PR [TRTLLM-14287][feat] Qwen Image CFG parallelism support #16384 (commit c083da6, "Qwen Image CFG parallelism") renamed QwenImagePipeline.forward()'s CFG kwarg true_cfg_scale → negative_prompt_cfg_scale and updated the three unit-test files, but left the integration LPIPS call site on the old name, so forward() raised TypeError: unexpected keyword argument 'true_cfg_scale'.
• Fix: Renamed the integration test's kwarg and its constant to negative_prompt_cfg_scale / QWENIMAGE_LPIPS_NEGATIVE_PROMPT_CFG_SCALE, corrected the stale explanatory comment, and removed the two now-obsolete waives.txt entries for this bug (the Qwen-Image-Layered call site at line 980 correctly keeps true_cfg_scale, since that pipeline's signature is unchanged).
• Automated fix generated by repair-bot

Test plan

• Verify fix on the same GPU type as the original failure
• Check for regressions in related tests

Links

• Bug: https://nvbugs/6550126

Dev Engineer Review

• Renamed the Qwen Image LPIPS argument and constant from true_cfg_scale to negative_prompt_cfg_scale.
• Updated the activation comment to match QwenImagePipeline.forward().
• Preserved true_cfg_scale for Qwen-Image-Layered because its API is unchanged.
• Removed two obsolete QwenImage entries from waives.txt.
• The affected LPIPS tests previously raised TypeError. The numerical result remains 0.009006.
• The changes are consistent with the updated API and do not change test scope or performance.

QA Engineer Review

• Modified the Qwen Image LPIPS test configuration in tests/integration/defs/examples/visual_gen/test_visual_gen.py.
• The CUDA-graph and LPIPS golden-comparison waiver entries were removed from tests/integration/test_lists/waives.txt.
• No test functions were added, modified, or removed.
• The affected tests are no longer waived and are covered by the existing visual-generation test configuration.
• Verdict: sufficient.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: bac48abc-8c10-43ce-a3b3-f5ecc876a8bd

📥 Commits

Reviewing files that changed from the base of the PR and between bbadfb7 and ed1301c.

📒 Files selected for processing (2)

• tests/integration/defs/examples/visual_gen/test_visual_gen.py
• tests/integration/test_lists/waives.txt

💤 Files with no reviewable changes (1)

• tests/integration/test_lists/waives.txt

---

Walkthrough

The QwenImage LPIPS visual-generation test now uses negative_prompt_cfg_scale. Its documentation and constant name were updated. Waivers for the CUDA-graph and standard LPIPS tests were removed.

Changes

QwenImage LPIPS validation

Layer / File(s)	Summary
Update CFG configuration and enable tests tests/integration/defs/examples/visual_gen/test_visual_gen.py, tests/integration/test_lists/waives.txt	The test replaces true_cfg_scale with negative_prompt_cfg_scale and renames the related constant. The waiver entries for the CUDA-graph and standard LPIPS tests were removed.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#17198: Removes other obsolete entries from tests/integration/test_lists/waives.txt.

Suggested reviewers: schetlur-nv, brnguyen2

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly identifies the bug fix and the renamed integration test keyword argument and constant.
Description check	✅ Passed	The description explains the root cause, fix, affected call sites, bug reference, and test plan; it omits the repository checklist.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[yibinl-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #63539 [ run ] triggered by Bot. Commit: ed1301c Link to invocation

[tensorrt-cicd]

PR_Github #63539 [ run ] completed with state SUCCESS. Commit: ed1301c
/LLM/main/L0_MergeRequest_PR pipeline #51505 completed with status: 'SUCCESS'

CI Report

Link to invocation

[StanleySun639]

LGTM if the target test case can pass.

[yibinl-nvidia]

LGTM