url/SkillOpt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: rename remaining teacher/student refs, remove .gradio from repo

Browse Source 
- Fix teacher/student in deep_reflect, meta_reflect, sealqa, babyvision,
  mathverse, mmrb, swebench envs and prompt templates
- Remove .gradio/certificate.pem from tracked files
- Add .gradio/ to .gitignore

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This commit is contained in:
Cuzyoung
2026-05-24 19:22:20 +00:00
parent 7ae2d8766e
commit cff7ff6846
27 changed files with 148 additions and 178 deletions
Download Patch File Download Diff File Expand all files Collapse all files
skillopt/envs/mathverse/adapter.py
+2 -2
View File
@@ -10,7 +10,7 @@ from skillopt.envs.mathverse.dataloader import MathVerseDataLoader
from skillopt.envs.mathverse.rollout import run_batch
from skillopt.gradient.deep_probe import generate_deep_probe_instruction
from skillopt.gradient.reflect import run_minibatch_reflect
from skillopt.model import get_student_backend
from skillopt.model import get_target_backend
class MathVerseAdapter(EnvAdapter):
@@ -176,7 +176,7 @@ class MathVerseAdapter(EnvAdapter):
selected_ids = {str(item["id"]) for item in selected_items}
selected_results = [row for row in results if str(row.get("id")) in selected_ids]
selected_examples = self.attach_reference_context(selected_results, selected_items)
codex_backend = get_student_backend() == "codex_exec"
codex_backend = get_target_backend() == "codex_exec"
if codex_backend:
selected_examples = self.attach_codex_probe_context(selected_examples, prediction_dir)
selected_metadata = []
skillopt/envs/mathverse/prompts/analyst_error.md
+2 -2
View File
@@ -1,7 +1,7 @@
You are an expert failure-analysis agent for visual mathematical reasoning problems.
You will be given MULTIPLE failed trajectories from a single minibatch and the current skill document.
Each trajectory includes the student's response, the evaluation result, and sometimes a hidden reference
Each trajectory includes the target's response, the evaluation result, and sometimes a hidden reference
containing the fuller Text Dominant version of the same problem.
Your job is to identify COMMON reasoning failures across the batch and propose concise skill edits.
@@ -17,7 +17,7 @@ Your job is to identify COMMON reasoning failures across the batch and propose c
1. Focus on patterns that recur across the minibatch.
2. Prefer edits that improve visual grounding and exact answer selection.
3. Do not hardcode problem-specific formulas or answers.
4. If hidden reference text is present, use it only to infer what information the student failed to recover from the Text Lite version.
4. If hidden reference text is present, use it only to infer what information the target failed to recover from the Text Lite version.
Respond ONLY with a valid JSON object:
{
skillopt/envs/mathverse/prompts/deep_probe.md
+5 -5
View File
@@ -1,16 +1,16 @@
You are an expert diagnostic-probe designer for visual mathematical reasoning tasks.
You will be shown representative trajectories, the current student skill, and the student's original prompt context.
You will be shown representative trajectories, the current target skill, and the target's original prompt context.
Some trajectories may also include a hidden reference containing the fuller Text Dominant wording of the same problem.
Design one SMALL diagnostic instruction that exposes the student's intermediate judgment without materially changing the original scaffold.
Design one SMALL diagnostic instruction that exposes the target's intermediate judgment without materially changing the original scaffold.
## Hard Constraints
1. Do NOT substantially change the original scaffold.
2. Do NOT prescribe a new long multi-step solving procedure.
3. Do NOT ask for a full proof or full chain-of-thought.
4. Ask only for a short readout of the signals already behind the student's current answer.
4. Ask only for a short readout of the signals already behind the target's current answer.
5. Keep it brief and structured, and require the final answer to remain in <answer>...</answer>.
6. If hidden reference text is present, use it only to target what visual or textual constraint the student likely missed.
6. If hidden reference text is present, use it only to target what visual or textual constraint the target likely missed.
## Good Probe Targets
- decisive diagram cue
@@ -21,5 +21,5 @@ Design one SMALL diagnostic instruction that exposes the student's intermediate
Respond ONLY with a valid JSON object:
{
"reasoning": "<why this probe is informative>",
"probe_instruction": "<the exact instruction text to append to the student prompt>"
"probe_instruction": "<the exact instruction text to append to the target prompt>"
}
skillopt/envs/mathverse/rollout.py
+12 -12
View File
@@ -8,8 +8,8 @@ import os
from concurrent.futures import ThreadPoolExecutor, as_completed
from skillopt.envs.mathverse.evaluator import evaluate_item, evaluation_mode, extract_answer
from skillopt.model import chat_student_messages, get_student_backend, is_student_exec_backend
from skillopt.model.codex_harness import prepare_workspace, render_skill_md, run_student_exec
from skillopt.model import chat_target_messages, get_target_backend, is_target_exec_backend
from skillopt.model.codex_harness import prepare_workspace, render_skill_md, run_target_exec
from skillopt.prompts import load_prompt
@@ -144,10 +144,10 @@ def _run_codex_once(
images=[item["image_path"]],
)
prompt = (
"Use the `skillopt-student` skill available in this workspace.\n"
"Use the `skillopt-target` skill available in this workspace.\n"
"Read `task.md`, inspect the attached image, solve the problem, and return only the final answer inside <answer>...</answer>."
)
final_message, raw = run_student_exec(
final_message, raw = run_target_exec(
work_dir=work_dir,
prompt=prompt,
model=model,
@@ -201,7 +201,7 @@ def process_one(
pred_dir = os.path.join(out_root, "predictions", item_id)
os.makedirs(pred_dir, exist_ok=True)
if is_student_exec_backend():
if is_target_exec_backend():
from skillopt.model import azure_openai as _llm
response = ""
@@ -215,7 +215,7 @@ def process_one(
pred_dir=pred_dir,
item=item,
skill_content=skill_content,
model=_llm.STUDENT_DEPLOYMENT,
model=_llm.TARGET_DEPLOYMENT,
timeout=120,
image_detail=image_detail,
diagnostic_mode=diagnostic_mode if turn == 0 else False,
@@ -230,9 +230,9 @@ def process_one(
result["response"] = response
result["agent_ok"] = True
result["n_turns"] = len(conversation) - 1
with open(os.path.join(pred_dir, "student_system_prompt.txt"), "w", encoding="utf-8") as f:
with open(os.path.join(pred_dir, "target_system_prompt.txt"), "w", encoding="utf-8") as f:
f.write(system_prompt)
with open(os.path.join(pred_dir, "student_user_prompt.txt"), "w", encoding="utf-8") as f:
with open(os.path.join(pred_dir, "target_user_prompt.txt"), "w", encoding="utf-8") as f:
f.write(user_text)
else:
messages, system_prompt, user_text = _build_messages(
@@ -249,7 +249,7 @@ def process_one(
]
for turn in range(max_turns):
if turn == 0:
resp_text, _ = chat_student_messages(
resp_text, _ = chat_target_messages(
messages=messages,
max_completion_tokens=1024,
retries=5,
@@ -267,7 +267,7 @@ def process_one(
{"role": "assistant", "content": response},
{"role": "user", "content": refinement_text},
]
resp_text, _ = chat_student_messages(
resp_text, _ = chat_target_messages(
messages=refinement_messages,
max_completion_tokens=768,
retries=5,
@@ -281,9 +281,9 @@ def process_one(
result["response"] = response
result["agent_ok"] = True
result["n_turns"] = len(conversation) - 1
with open(os.path.join(pred_dir, "student_system_prompt.txt"), "w", encoding="utf-8") as f:
with open(os.path.join(pred_dir, "target_system_prompt.txt"), "w", encoding="utf-8") as f:
f.write(system_prompt)
with open(os.path.join(pred_dir, "student_user_prompt.txt"), "w", encoding="utf-8") as f:
with open(os.path.join(pred_dir, "target_user_prompt.txt"), "w", encoding="utf-8") as f:
f.write(user_text)
eval_result = evaluate_item(