eslbench-report-analysis

name	eslbench-report-analysis
description	Use when analyzing ESL-Bench evaluation reports - extracting scores by difficulty dimension, comparing methods, checking fail rates, token costs, duration per query. Triggers on keywords like eslbench report, benchmark results, group by difficulty.

ESL-Bench Report Analysis

Extract and compare evaluation results from HolyEval benchmark report JSON files.

Report Locations

~/xiaotong/holyeval/benchmark/report/eslbench/
  {dataset}_{target_type}_{model}_{date}_{time}.json
~/*/holyeval/benchmark/report/eslbench/
  {dataset}_{target_type}_{model}_{date}_{time}.json

Example report paths for recent 3 days: YYYYMMDD-3 to YYYYMMDD: ~/xiaotong/holyeval/benchmark/report/eslbench/sample50-20260331_*_{YYYYMMDD-2,YYYYMMDD-1,YYYYMMDD}*.json

Report JSON Structure

report = {
    "benchmark_name": "eslbench",
    "dataset_name": "sample50-20260331",
    "runtime_target": {...},
    "cases": [...],           # Per-case results
    "pass_count": 0,          # Cases scoring >= threshold
    "fail_count": 3,          # Cases with eval errors (not low scores)
    "pass_rate": 0.0,         # pass_count / total (NOTE: often 0 due to strict threshold)
    "avg_score": 0.512,       # Mean score across all cases (0-1 scale)
    "total_duration_seconds": 8492
}

Per-Case Structure

case = {
    "id": "user5022_AT_demo_Q001",
    "title": "[Lookup] ...",
    "tags": ["difficulty:Trend", "answer_type:numeric_value"],
    "eval": {
        "result": "scored",    # or "error"
        "score": 0.85,         # 0.0 - 1.0
        "feedback": "..."
    },
    "cost": {
        "target": {
            "gpt-5.4": {
                "input_tokens": 8621,
                "output_tokens": 314,
                "total_tokens": 8935,
                "input_token_details": {"cache_read": 7168},
                "output_token_details": {"reasoning": 0}
            }
        }
    },
    "start": "...", "end": "..."
}

Key Fields

Field	Where	Notes
Difficulty dimension	case.tags → "difficulty:Lookup"	5 values: Lookup, Trend, Comparison, Anomaly, Explanation
Score	case.eval.score	0.0-1.0, multiply by 100 for percentage
Answer type	case.tags → "answer_type:text"	text, numeric_value, list, boolean
Fail	report.fail_count	Eval errors, not low scores
Token usage	case.cost.target.{model}.input_tokens	May be None/empty if not tracked
Duration	report.total_duration_seconds	Wall-clock for entire run

Resolving Input Files

Input can be any of:

• Glob pattern: ~/xiaotong/holyeval/benchmark/report/eslbench/sample50-20260331_*.json
• Directory: scan all *.json inside
• Explicit file list: one or more JSON file paths
• Zip URL: download, extract, scan *.json inside
• Zip file path: extract, scan *.json inside

The analysis output MUST show the full resolved file path for each report processed.

CRITICAL: Script stdout may be truncated when output is large (>2KB). To ensure file paths are always visible to the user:

1. Write the resolved file path list into the markdown report file (under a ## Processed Report Files section)
2. Also print a brief summary (file count + sources) directly to the user in your text response, e.g. "31 files analyzed: 21 LOCAL + 10 ZIP"

Do NOT rely on script stdout alone for showing file paths — it will be truncated for large analyses.

Default report location (if no input specified): ~/xiaotong/holyeval/benchmark/report/eslbench/

Analysis Script Template

import json, glob, os, re, zipfile, tempfile
from collections import defaultdict

DIMS = ["Lookup", "Trend", "Comparison", "Anomaly", "Explanation"]

def resolve_report_files(inputs):
    """Resolve various input forms to a sorted list of JSON file paths.
    inputs: list of paths, glob patterns, directories, or zip files."""
    files = []
    for inp in inputs:
        inp = os.path.expanduser(inp)
        if inp.endswith(".zip") and os.path.isfile(inp):
            tmpdir = tempfile.mkdtemp()
            with zipfile.ZipFile(inp) as z:
                z.extractall(tmpdir)
            files.extend(sorted(glob.glob(f"{tmpdir}/**/*.json", recursive=True)))
        elif os.path.isdir(inp):
            files.extend(sorted(glob.glob(f"{inp}/**/*.json", recursive=True)))
        elif "*" in inp or "?" in inp:
            files.extend(sorted(glob.glob(inp)))
        elif os.path.isfile(inp):
            files.append(inp)
    # Filter out __MACOSX etc
    return [f for f in files if "/__MACOSX/" not in f and f.endswith()]

 ():
     inputs  :
        inputs = [os.path.expanduser()]
    files = resolve_report_files(inputs)
    rows = []
     fpath  files:
        fname = os.path.basename(fpath)
         (fpath)  f:
            report = json.load(f)
        cases = report.get(, [])
        n = (cases)
         n == : 
        fail_n = report.get(, )
        dur = report.get(, )

        by_diff = defaultdict()
        in_tok = out_tok = 
         c  cases:
            diff = ((t.split(,)[]  t  c.get(,[])  t.startswith()), )
            score = c.get(,{}).get()
             diff  score   :
                by_diff[diff].append(score)
             model_usage  (c.get(,{}).get()  {}).values():
                 (model_usage, ):
                    in_tok += model_usage.get(,)  
                    out_tok += model_usage.get(,)  

        all_scores = [s  scores  by_diff.values()  s  scores]
        rows.append({
            : fpath,  
            : fname,  
            : n,
            : {d: ((by_diff[d])/(by_diff[d])*  by_diff[d]  )  d  DIMS},
            : (all_scores)/(all_scores)*  all_scores  ,
            : , : fail_n,
            : dur,
            : dur/n  n  ,
            : in_tok, : out_tok,
        })
     rows

 ():
    
     n == :  
     n >= :  
     n >= :  
     (n)

 ():
    
    
    ()
     i, r  (rows, ):
        ()
    ()
    
     ():    v     
    ()
    ()
     r  rows:
        d = r[]
        ()

After Score Overview: Write a Summary

After printing the score table, always write a free-form summary covering:

1. Best overall performer — which report/model has highest Avg%, noting fail rate trade-off
2. Per-dimension winners — which model leads each dimension (Look/Trend/Comp/Anom/Expl)
3. Weakest dimension — which dimension is universally low across all methods and why
4. Efficiency — token usage vs. score (cost-effectiveness), duration per query
5. Reliability — fail rates, runs that should be discarded (high fail = unreliable)
6. Notable surprises — unexpected results (e.g. structured API outperforming LLM, score jumps between runs)

Format as a ## Summary section with bullet points. Be specific with numbers, don't just say "good" or "bad".

Fail Reason Analysis

Cases with eval.result == "fail" are execution errors, not low scores. Classify by extracting error from eval.feedback traceback.

from collections import Counter

def analyze_fails(report):
    """Classify fail reasons from a report. Returns Counter of reason -> count."""
    reasons = Counter()
    for c in report.get("cases", []):
        ev = c.get("eval", {})
        if ev.get("result") != "fail":
            continue
        fb = ev.get("feedback", "")
        if "TimeoutError" in fb or "CancelledError" in fb:
            reasons["Timeout/CancelledError"] += 1
        elif "orchestrator.py" in fb:
            reasons["Orchestrator error (model API failure)"] += 1
        elif "asyncio/tasks.py" in fb:
            reasons["Async task error"] += 1
        else:
            reasons[fb[:80] if fb else "unknown"] += 1
    return reasons

Common Fail Categories

Category	Cause	Fix
Timeout/CancelledError	Query took too long, asyncio cancelled	Increase timeout or reduce concurrency
Orchestrator model error	Target model API returned error (rate limit, bad response)	Check API key/quota, retry
Async task error	General asyncio failure	Check logs for root cause

When to Run Fail Reason Analysis

For every report with fail_count > 0, run analyze_fails() and include the breakdown in:

1. The markdown report — add a ## Fail Reason Breakdown section listing each report's fail sub-categories (Timeout vs Orchestrator vs Async vs other)
2. The per-dimension error analysis — when showing FAIL_execution_error:N, also note the sub-category if available (e.g. FAIL_execution_error:3 (2 Timeout, 1 Orchestrator))

Do NOT just report the total FAIL_execution_error count without sub-classifying.

Per-Agent Per-Dimension Error Analysis

Classify WHY each agent scores low on each dimension. Combines fail cases and low-score cases into one view.

Feedback Classification

eval.feedback format: [answer_type] score=X.XXX — detail

Category	Pattern in feedback	Meaning
no_data_found	"no_data", "no relevant"	Agent couldn't retrieve data
numeric_mismatch	"Number match 0/"	Computed number wrong
partial_list_match	"F1="	Incomplete list returned
rubric_low_score	"LLM judge", "Rubric"	LLM judge scored low on rubric
exact_match_fail	"exact" + "0/"	Text/boolean exact match failed
FAIL_execution_error	eval.result == "fail"	Execution error (timeout/API)

Analysis Script

import json, glob, os, re
from collections import Counter

REPORT_DIR = os.path.expanduser("~/xiaotong/holyeval/benchmark/report/eslbench")
DIMS = ["Lookup", "Trend", "Comparison", "Anomaly", "Explanation"]

def classify_feedback(fb):
    """Classify a single eval.feedback string into an error category."""
    m = re.match(r'\[(\w+)\]\s*score=([\d.]+)\s*[—-]\s*(.*)', fb)
    if not m:
        return fb[:50] or "unknown"
    _, _, detail = m.groups()
    dl = detail.lower()
    if "no_data" in dl or "no relevant" in dl:
        return "no_data_found"
    if "number match 0/" in dl:
        return "numeric_mismatch"
    if "f1=" in dl:
        return "partial_list_match"
    if "llm judge" in dl or "rubric" in dl:
        return "rubric_low_score"
    if "exact" in dl    dl:
         
     detail[:]

 ():
    
    results = {}
     dim  DIMS:
        reasons = Counter()
        reason_cases = defaultdict()  
        scores = []
         c  report.get(, []):
            diff = ((t.split(,)[]  t  c.get(,[])
                          t.startswith()), )
             diff != dim:
                
            ev = c.get(, {})
            cid = c.get(, )
             ev.get() == :
                reasons[] += 
                reason_cases[].append(cid)
                
            score = ev.get()
             score  :
                
            scores.append(score)
             score < threshold:
                cat = classify_feedback(ev.get(, ))
                reasons[cat] += 
                reason_cases[cat].append(cid)
        results[dim] = {
            : (scores)/(scores)*  scores  ,
            : (scores),
            : (reasons.values()),
            : reasons,
            : reason_cases,
        }
     results

 ():
    
     inputs  :
        inputs = [os.path.expanduser()]
    files = resolve_report_files(inputs)
     fpath  files:
        fname = os.path.basename(fpath)
         (fpath)  f:
            report = json.load(f)
        fail_n = report.get(, )
        n = (report.get(, []))
        ()
        ()
        ()
        ()
        results = analyze_report_by_dim(report)
         dim  DIMS:
            r = results[dim]
            top3 = r[].most_common()
            top_str = .join(  k,v  top3)  top3  
            ()
             show_cases > :
                 reason, count  top3:
                    ids = r[][reason][:show_cases]
                    suffix =   count > show_cases  
                    ()

Cross-Agent Dimension Patterns

When multiple agent types are present (LLM-direct, RAG, Structured API / theta), always generate a cross-agent comparison table in the markdown report under ## Cross-Agent Error Patterns. Group agents into categories and show the top 2 error types per dimension per category.

Template (populate from actual data, not hardcoded):

Dimension	LLM-direct (top errors)	RAG (top errors)	Structured API (top errors)
Lookup	...	...	...
Trend	...	...	...
Comparison	...	...	...
Anomaly	...	...	...
Explanation	...	...	...

Agent category mapping:

• LLM-direct: llm_api target type (gpt-5.4, gemini-flash, claude-sonnet, minimax, glm, etc.)
• RAG: *_rag_api target types (hippo_rag, dyg_rag, mem0_rag, evermem)
• Structured API: theta_api, theta_smart_api

After the table, add a Key insights paragraph noting which categories dominate which dimensions and why.

Reference patterns (from prior analyses — verify against current data):

• LLM-direct tends to fail on Comparison (partial_list_match) and Explanation (numeric_mismatch)
• RAG methods suffer numeric_mismatch across all dimensions — retrieval noise degrades computation
• Structured API at 0% may be connectivity issue — check if all scores are exactly 0.0
• Comparison is typically the weakest dimension across all agent types
• Anomaly discriminates most: LLM-direct scores 72-98% while RAG drops to 1-48%

Error Sample Cases Export

When running analysis, always export error/low-score cases to a JSON file alongside the markdown report. This enables case-level debugging without re-parsing report files.

Export Script

def export_error_cases(inputs=None, threshold=0.8, output_path=None):
    """Export all low-score and failed cases to a JSON file for debugging.
    
    Args:
        inputs: same as analyze_reports()
        threshold: score below this is considered error (default 0.8)
        output_path: output JSON path. Default: docs/eslbench_error_cases-YYYYMMDD.json
    
    Returns: output file path
    """
    import datetime
    if inputs is None:
        inputs = [os.path.expanduser("~/xiaotong/holyeval/benchmark/report/eslbench")]
    if output_path is None:
        today = datetime.date.today().strftime("%Y%m%d")
        output_path = f"docs/eslbench_error_cases-{today}.json"

    files = resolve_report_files(inputs)
    error_cases = []

    for fpath in files:
        fname = os.path.basename(fpath)
        with open(fpath) as f:
            report = json.load(f)
        for c in report.get("cases", []):
            ev = c.get("eval", {})
            score = ev.get("score")
            result = ev.get("result", "")
            is_fail = result in ("fail", "error")
            is_low = score is not None and score < threshold

            if not is_fail and not is_low:
                

            diff = ((t.split(,)[]  t  c.get(,[])
                          t.startswith()), )
            atype = ((t.split(,)[]  t  c.get(,[])
                           t.startswith()), )
            category =   is_fail  classify_feedback(ev.get(, ))

            error_cases.append({
                : fname,
                : c.get(, ),
                : diff,
                : atype,
                : score,
                : result,
                : category,
                : ev.get(, ),
                : c.get(, ),
                : c.get(, ),
            })

    
    diff_order = {d: i  i, d  (DIMS)}
    error_cases.sort(key= x: (
        x[],
        diff_order.get(x[], ),
        x[]  x[]     -,
    ))

    os.makedirs(os.path.dirname(output_path)  , exist_ok=)
     (output_path, )  f:
        json.dump(error_cases, f, indent=, ensure_ascii=)

    
     collections  Counter
    by_cat = Counter(c[]  c  error_cases)
    by_diff = Counter(c[]  c  error_cases)
    ()
    ()
    ()
     output_path

Output Format

Each exported case contains:

Field	Description
report_file	Source report filename
case_id	e.g. user5022_AT_demo_Q001
difficulty	Lookup/Trend/Comparison/Anomaly/Explanation
answer_type	text/numeric_value/list/boolean
score	0.0-1.0, null for execution errors
eval_result	"scored", "fail", "error"
error_category	Classified: numeric_mismatch, partial_list_match, etc.
feedback	Full eval feedback string
title	Question title
description	Full question description

When to Export

Always call export_error_cases() after print_table() and print_full_analysis(). The exported file is the entry point for case-level debugging — use it to trace specific failures back to dataset issues or model weaknesses.

Interpretation Notes

• pass_rate: Often 0.0 because it uses a strict binary threshold; use avg_score instead for meaningful comparison
• fail_count: Eval-level errors (API timeout, parse failure), NOT low-scoring answers. High fail = unreliable run
• theta_api_expert at 0%: Usually means API connectivity issue, not model failure
• Token cost = 0: Means token tracking not enabled for that target type (e.g. mem0_rag, theta_api)
• Multiple runs of same model: Take the latest or lowest-fail run; early runs may have config bugs
• s/q (seconds per query): Dominated by retrieval + generation time; <30s usually means cached/fast path

Per-Agent Performance Analysis (Verified Runs Only)

Filter criteria: Only include runs where fail_rate < 40% AND scored_count > 30 (i.e. at least 30 successfully scored answers). This excludes partial/broken runs and ensures statistically meaningful comparisons.

When multiple runs of the same agent exist, keep only the latest run that passes the filter (by timestamp in filename).

Agent Performance Table

Generate a ## Agent Performance (Verified) section with a filtered, deduplicated table:

def get_verified_agents(rows):
    """Filter to verified runs: fail_rate < 40% and scored > 30. Deduplicate by agent, keep latest."""
    verified = []
    for r in rows:
        scored = r["n"] - r["fail_n"]
        fail_rate = r["fail_n"] / r["n"] if r["n"] else 1.0
        if fail_rate < 0.40 and scored > 30:
            verified.append(r)

    # Deduplicate: extract agent key (target_type + model), keep latest by filename timestamp
    from collections import OrderedDict
    by_agent = OrderedDict()
    for r in sorted(verified, key=lambda x: x["label"]):
        # Extract agent key: everything before the date_time suffix
        # e.g. "sample50-20260407_llm_api_gpt-5.4" from "sample50-20260407_llm_api_gpt-5.4_20260407_190032.json"
        parts = r["label"].rsplit("_", 2)  # split off YYYYMMDD_HHMMSS.json
        agent_key = parts[0] if len(parts) >= 3 else r["label"]
        if agent_key not in by_agent or r["label"] > by_agent[agent_key]["label"]:
            by_agent[agent_key] = r
    return (by_agent.values())

Table Format

| Agent | Look | Trend | Comp | Anom | Expl | Avg% | Scored | Fail | s/q | InTok |

Sort by Avg% descending. This is the primary comparison table — the Score Overview shows all raw runs, this shows only verified, deduplicated results.

Per-Agent Dimension Radar Summary

For each verified agent, add a one-line text radar showing relative strengths:

gpt-5.4: Trend(70.0) > Anom(57.1) > Comp(44.4) > Look(30.1) > Expl(14.2)

This helps quickly identify each agent's strength/weakness profile.

Output to File

When saving analysis, produce two files:

1. Markdown report: docs/eslbench_report_analysis-YYYYMMDD.md — must contain ALL of:
   • ## Processed Report Files — numbered list of full resolved file paths (with source tag if mixed)
   • ## Score Overview — the score table
   • ## Summary — the 6-point free-form analysis
   • ## Fail Reason Breakdown — per-report sub-classification of fails (Timeout/Orchestrator/Async/other), only for reports with fail_count > 0
   • ## Cross-Agent Error Patterns — the cross-agent dimension comparison table (if multiple agent types present)
   • ## Agent Performance (Verified) — deduplicated table of agents with fail_rate < 40% and scored > 30, plus per-agent dimension radar
   • ## Error Distribution — aggregate error category and difficulty counts
2. Error cases JSON: docs/eslbench_error_cases-YYYYMMDD.json — all low-score/failed cases for debugging

Always print the mdpreview link for the markdown report:

http://10.241.13.122:22086/preview.html?path={path-relative-to-/home/fat/}

Example: if saved to /home/fat/caill/thetagendata/docs/eslbench_report_analysis-20260401.md, output:

http://10.241.13.122:22086/preview.html?path=caill/thetagendata/docs/eslbench_report_analysis-20260401.md
Error cases: docs/eslbench_error_cases-20260401.json (N cases)

Final: Verify Output Sections

After writing the markdown report, MUST run this verification before completing:

REQUIRED_SECTIONS=("## Processed Report Files" "## Score Overview" "## Summary" "## Fail Reason Breakdown" "## Cross-Agent Error Patterns" "## Agent Performance (Verified)" "## Error Distribution")
MISSING=()
for section in "${REQUIRED_SECTIONS[@]}"; do
  grep -qF "$section" "$OUTPUT_FILE" || MISSING+=("$section")
done
if [ ${#MISSING[@]} -eq 0 ]; then
  echo "OK: All 6 required sections present"
else
  echo "MISSING sections:"
  printf "  - %s\n" "${MISSING[@]}"
fi

If any section is MISSING, add it to the report file before completing. Exceptions:

• ## Fail Reason Breakdown may be omitted ONLY if ALL reports have fail_count == 0
• ## Cross-Agent Error Patterns may be omitted ONLY if all reports are from the same agent type
• ## Agent Performance (Verified) may be omitted ONLY if no runs pass the filter (fail_rate < 40% AND scored > 30)