SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.14735 · 2025

EAGER-LLM: Enhancing Large Language Models as Recommenders through Exogenous Behavior-Semantic Integration

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 3 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2502.14735.json · badge: /api/ran/2502.14735.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Indolent-Kawhi/EAGER-LLM canonical (classifier label)
classifier basis not recorded on the link
9 ran · 1 no recorded run
FunctionStatusWhere it lives
atoi Ran
ran, honoured the contract we drafted
Indolent-Kawhi/EAGER-LLM/convert/zero_to_fp32.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from harshmaru7/DilatedConv, is pointer only (licence: NONE) · get_code("6dbf1b2a5b901370")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from harshmaru7/DilatedConv, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_numbers Ran
ran on a synthesized input
Indolent-Kawhi/EAGER-LLM/src/evaluate.py
code served (permissive licence) · get_code("fdaba9b1c31749ae")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_numbers(text):
    text_list = text.split('><')
    text_num = [0] * len(text_list)
    for i in range(len(text_list)):
        text_num[i] = int(re.search('\\d+', text_list[i]).group())
    return text_num

ARGS = {
    'text': '<item1><item2><item3>'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_model_state_file Ran
ran; our contract draft was wrong, not the code
Indolent-Kawhi/EAGER-LLM/convert/zero_to_fp32.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from rucaibox/lc-rec, is pointer only (licence: NONE) · get_code("3a53f3109825e747")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from rucaibox/lc-rec, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_topk_results Ran
ran on a synthesized input
Indolent-Kawhi/EAGER-LLM/src/evaluate.py
code served (permissive licence) · get_code("635647232f70b48c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_numbers(text):
    text_list = text.split('><')
    text_num = [0] * len(text_list)
    for i in range(len(text_list)):
        text_num[i] = int(re.search('\\d+', text_list[i]).group())
    return text_num

def match_codes(prediction, target_item):
    if prediction == target_item:
        return True
    if prediction.find('unk') != -1:
        pred_num = extract_numbers(prediction)
        target_num = extract_numbers(target_item)
        cnt = 0
        for i in range(len(pred_num)):
            if pred_num[i] != target_num[i]:
                cnt += 1
        if cnt <= 1:
            print('prediction:{} target_item:{}'.format(prediction, target_item))
            return True
    return False

def get_topk_results(predictions, scores, targets, k, all_items=None):
    results = []
    B = len(targets)
    predictions = [_.split('Response:')[-1] for _ in predictions]
    predictions = [_.strip().replace(' ', '') for _ in predictions]
    wrong_code = 0
    if all_items is not None:
        for i, seq in enumerate(predictions):
            if seq not in all_items:
                scores[i] = -1000
                wrong_code += 1
    for b in range(B):
        batch_seqs = predictions[b * k:(b + 1) * k]
        batch_scores = scores[b * k:(b + 1) * k]
        pairs = [(a, b) for a, b in zip(batch_seqs, batch_scores)]
        sorted_pairs = sorted(pairs, key=lambda x: x[1], reverse=True)
        target_item = targets[b]
        one_results = []
        for sorted_pred in sorted_pairs:
            if match_codes(sorted_pred[0], target_item) and sorted_pred[1] > -1000:
                one_results.append(1)
            else:
                one_results.append(0)
        results.append(one_results)
    return (results, wrong_code)

ARGS = {
    "predictions": [
        "Response: code1>1<code2>2<",
        "Response: code1>1<code2>2<",
        "Response: code1>1<code2>3<",
        "Response: code1>1<code2>3<",
    ],
    "scores": [0.9, 0.8, 0.7, 0.6],
    "targets": [
        "code1>1<code2>2<",
        "code1>1<code2>3<",
    ],
    "k": 2,
    "all_items": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_reviews Ran
ran on a synthesized input
Indolent-Kawhi/EAGER-LLM/src/data_process/amazon14_data_process.py
code served (permissive licence) · get_code("10f2f49dffdbfeae")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import gzip
import tempfile
import os

# Create a temporary gzip file with mock review data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "reviews.json.gz")

# Write mock review data in the format expected by parse()
mock_reviews = [
    b"{'reviewerID': 'A1', 'asin': 'B1', 'overall': 5.0, 'unixReviewTime': 1000000}",
    b"{'reviewerID': 'A2', 'asin': 'B2', 'overall': 4.0, 'unixReviewTime': 1000001}",
    b"{'reviewerID': 'A1', 'asin': 'B2', 'overall': 3.0, 'unixReviewTime': 1000002}",
    b"{'reviewerID': 'A3', 'asin': 'B1', 'overall': 2.0, 'unixReviewTime': 1000003}",
]

with gzip.open(temp_file, 'wb') as f:
    for review in mock_reviews:
        f.write(review + b'\n')

ARGS = {
    'file': temp_file,
    'rating_score': 0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
match_codes Ran
ran on a synthesized input
Indolent-Kawhi/EAGER-LLM/src/evaluate.py
code served (permissive licence) · get_code("0faf132067ed08d6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_numbers(text):
    text_list = text.split('><')
    text_num = [0] * len(text_list)
    for i in range(len(text_list)):
        text_num[i] = int(re.search('\\d+', text_list[i]).group())
    return text_num

def match_codes(prediction, target_item):
    if prediction == target_item:
        return True
    if prediction.find('unk') != -1:
        pred_num = extract_numbers(prediction)
        target_num = extract_numbers(target_item)
        cnt = 0
        for i in range(len(pred_num)):
            if pred_num[i] != target_num[i]:
                cnt += 1
        if cnt <= 1:
            print('prediction:{} target_item:{}'.format(prediction, target_item))
            return True
    return False

ARGS = {
    'prediction': 'code1><unk2><code3',
    'target_item': 'code1><code2><code3'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
natural_keys Ran
ran; our contract draft was wrong, not the code
Indolent-Kawhi/EAGER-LLM/convert/zero_to_fp32.py
code served (permissive licence) · get_code("7052cf0436949e37")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def atoi(text):
    return int(text) if text.isdigit() else text

def natural_keys(text):
    """
    alist.sort(key=natural_keys) sorts in human order
    http://nedbatchelder.com/blog/200712/human_sorting.html
    (See Toothy's implementation in the comments)
    """
    return [atoi(c) for c in re.split('(\\d+)', text)]

# The function takes a single parameter: text (a string)
# It uses string methods: isdigit(), re.split()
# It returns a list of mixed ints and strings

ARGS = {
    "text": "file123name456.txt"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
parse_dataset_args Ran
ran on a synthesized input
Indolent-Kawhi/EAGER-LLM/src/utils.py
code served (permissive licence) · get_code("8572253beca3ba2b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArgumentParser. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import argparse
import torch
import numpy as np

torch.manual_seed(0)

ARGS = {
    'parser': argparse.ArgumentParser(),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
parse_global_args Ran
ran; our contract draft was wrong, not the code
Indolent-Kawhi/EAGER-LLM/src/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from honghuibao2000/letter, is pointer only (licence: NONE) · get_code("2be50b7fb0f78745")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from honghuibao2000/letter, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

parse_train_args Not yet run
unverified (harvested, no recorded run)
Indolent-Kawhi/EAGER-LLM/src/utils.py
code served (permissive licence) · get_code("c89eceaa0a5268a8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 3 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
atoi 13 other papers
12 dated earlier, 1 later
1511.07122 (2015-11)
Multi-Scale Context Aggregation by Dilated Convolutions
1908.03706 (2019-08)
Exploiting temporal consistency for real-time video depth estimation
2104.08773 (2021-04)
Cross-Task Generalization via Natural Language Crowdsourcing Instructions
and 10 more not named here
natural_keys 12 other papers
11 dated earlier, 1 later
1511.07122 (2015-11)
Multi-Scale Context Aggregation by Dilated Convolutions
1908.03706 (2019-08)
Exploiting temporal consistency for real-time video depth estimation
2203.03339 (2022-03)
L2CS-Net: Fine-Grained Gaze Estimation in Unconstrained Environments
and 9 more not named here
get_model_state_file 5 other papers
4 dated earlier, 1 later
2308.14034 (2023-08)
Confucius: Iterative Tool Learning from Introspection Feedback by Easy-to-Difficult Curriculum
2310.09168 (2023-10)
EXPLORE-INSTRUCT: Enhancing Domain-Specific Instruction Coverage through Active Exploration
2311.04072 (2023-11)
Beyond Imitation: Leveraging Fine-grained Quality Signals for Alignment
and 2 more not named here
parse_global_args 2 other papers
2 dated earlier
2311.09049 (2023-11)
Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation
2405.07314 (2024-05)
Learnable Item Tokenization for Generative Recommendation

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Large language models (LLMs) are increasingly leveraged as foundational backbones in the development of advanced recommender systems, offering enhanced capabilities through their extensive knowledge and reasoning. Existing llm-based recommender systems (RSs) often face challenges due to the significant differences between the linguistic semantics of pre-trained LLMs and the collaborative semantics essential for RSs. These systems use pre-trained linguistic semantics but learn collaborative semantics from scratch via the llm-Backbone. However, LLMs are not designed for recommendations, leading to inefficient collaborative learning, weak result correlations, and poor integration of traditional RS features. To address these challenges, we propose EAGER-LLM, a decoder-only llm-based generative recommendation framework that integrates endogenous and exogenous behavioral and semantic information in a non-intrusive manner. Specifically, we propose 1)dual-source knowledge-rich item indices that integrates indexing sequences for exogenous signals, enabling efficient link-wide processing; 2)non-invasive multiscale alignment reconstruction tasks guide the model toward a deeper understanding of both collaborative and semantic signals; 3)an annealing adapter designed to finely balance the model's recommendation performance with its comprehension capabilities. We demonstrate EAGER-LLM's effectiveness through rigorous testing on three public benchmarks.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2502.14735")
get_code_for_paper("2502.14735")
have("2502.14735")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2502.14735.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2502.14735.svg)](https://syntology.ai/paper/2502.14735)

Connect an agent — have() is free.