FlowingFlowing
← Back to all development cases
This case was judged by an anonymized AI reviewer and has not been independently reviewed by people; inspect the draft, cards, and both outputs to assess it.
AutocompleteWith cards better209 / 280 · beu-2ba30d8d2c783a03

Internet of Behaviors: A Survey

Psychology and cognitive neuroscience · 2211.15588v1

FLOWING EVIDENCE BENCHMARK

How do we compare writing with and without cards?

At the same writing position, the same model produces one first result with cards and one without. All 280 pairs enter the overall anonymized review, which assesses whether a decision not to continue fits the draft.

Overall: 138 with-card wins, 26 ties, 107 no-card wins, 5 pairs where neither was usable, and 4 inconclusive judgments. Both sides wrote text in 218 pairs; 48 had a deliberate stop and 14 included an invalid output.

SAME DRAFT · WITH OR WITHOUT CARDS

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED VERSIONS

SHARED BY BOTH

Draft, writing task, and generation model

A · With cards

Relevant literature cards supplied

B · Without cards

No literature cards supplied

Generation model

Same model and version

A → first output

B → first output

Anonymized AI judge

First outputs are anonymized as X and Y

Continuation review: accuracy, fit to the writing task and usability
Returns: X preferred / tie / Y preferred / both unusable

Review again with the answer order swapped

HOW DOES BLIND JUDGING WORK?

① Anonymize both outputs
The AI judge sees the same draft and both first outputs without knowing which received cards.

② Compare and swap order
The judge applies task-specific criteria in X/Y and then Y/X order.

③ Review disagreements
A third pass resolves disagreements. Incomplete reviews remain in the denominator.

How are positions and cards selected?

We selected 40 writing positions in each of seven fields, for 280 comparisons. The position and relevant literature cards were set before each comparison, so both versions began from the same draft.

The main result retains every frozen position, including places where cards might not add a useful next move. This avoids selecting cases based on their outputs, but 49% describes the whole cohort, not a win rate where evidence clearly had an opportunity to help. This cohort did not record that opportunity label for each position before generation, so those two groups cannot yet be separated reliably.

CARD USE · 140/280

Reported card use in continuations with text

The model reports whether a card affected its continuation. In 140 of 280 tests it produced continuation text and reported card use. The previous count of 150 included 10 cases without continuation text and has been corrected.

COMPARISON · 70 / 7 / 30

Which written version was better?

Of the 107 reported-use cases where both versions produced text, the blinded AI judge preferred the with-card version 70 times, found 7 about the same, and preferred the without-card version 30 times.

How are the two table percentages calculated?

ADOPTION COVERAGE

140reported use with text
÷
280all positions
=
50%

WITH-CARD VERSION PREFERRED

70with cards better
÷
107comparable tests
=
65%

The percentages use different groups: 50% counts continuations with text reporting card use among all 280 tests; 65% covers the 107 such cases with text on both sides. The generating model reports card use, while an anonymized AI judge compares writing quality.

Card use is self-reported by the generator. The anonymized AI judge sees the draft and source snippets, so source-aware preference remains possible; there was no independent human review. The overall result describes these 280 frozen positions, not a win probability for any writing task.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

Internet of Behaviors: A Survey

I Introduction

…. (3) We provide a comprehensive summary of the current applications of the Internet of Behavior, including specific discussions in five scenarios that give an overview of the application status of the Internet of Behavior. (4) We discuss the challenges of the Internet of Behavior’s development and its future directions, which hopefully will bring some progress to the Internet of Behavior. To the best of our knowledge, this is the first survey paper on the Internet of Behavior. We hope that this in-depth review can provide some useful directions for more productive research in related fields.

…nes to the Internet, so people could check for cold drinks before making a purchase. It is widely regarded as one of the first IoT devices. The IoT got its name in 1999. Kevin Ashton, head of MIT’s Automatic ID Lab, first coined the term “Internet of Things” in a speech to illustrate the potential of RFID tracking technology. In the decades since, devices connected to the IoT have evolved from mobile phones and computers to daily devices like refrigerators and vending machines, and now smart home devices like smoke alarms, speaker boxes, and tablet computers are all connected to the Internet.

↓ Next paragraph was generated here

The highlighted text is near the writing position. This is a layout of the evaluation input, not a PDF screenshot.

A matched comparison at the same writing positionOverall AI review: With cards better

Both arms use the same model, task and manuscript context. Only the retrieved passages differ. The first recorded output from each arm is shown below.

Read overall AI review rationale

A = without cards · B = with cards

Both outputs attempt to bridge from IoT history toward the paper's actual topic (Internet of Behavior). However, they serve different argumentative functions. Output A abruptly introduces IoB with 'extends this connectivity' without sufficient transition—the draft has not yet established what 'this connectivity' contextually means for behavioral data, and the leap from Coke machine/RFID history to 'understand and influence behavior' is conceptually jarring. The phrase 'rather than merely to connect objects' also creates a subtle tension with the preceding paragraph's celebratory tone about IoT expansion. Output B more organically continues the historical trajectory by noting how IoT proliferation into specific domains (healthcare, transportation, smart homes, agriculture) naturally produces data about human activity. This sets up the IoB transition more gracefully: it shows the *mechanism* by which IoT becomes relevant to behavior (data streams capturing human activity) without prematurely naming IoB. The scope is appropriately narrower—describing a precondition for IoB rather than defining IoB itself, which fits better at this early stage of the manuscript. Output B requires less repair to integrate; Output A would need additional sentences to justify its definitional leap.

Evidence

Same model and position · retrieved passages supplied · first output

Preferred by blind review

This proliferation of connected devices has driven the IoT into domains such as healthcare, transportation, smart homes, and agriculture, where the resulting data streams increasingly capture not only machine states but human activity and behaviour.

Passages supplied to the Evidence version

Card 1 · RFID Localisation for Internet of Things Smart Homes : A Survey

The Internet of Things (IoT) enables numerous business opportunities in fields as diverse as e-health, smart cities, smart homes, among many others.The IoT incorporates multiple long-range, short-range, and personal area wireless networks and technologies into the designs of IoT applications.Localis…

Read full passage excerpt

The Internet of Things (IoT) enables numerous business opportunities in fields as diverse as e-health, smart cities, smart homes, among many others.The IoT incorporates multiple long-range, short-range, and personal area wireless networks and technologies into the designs of IoT applications.Localisation in indoor positioning systems plays an important role in the IoT.Location Based IoT applications range from tracking objects and people in real-time, assets management, agriculture, assisted monitoring technologies for healthcare, and smart homes, to name a few.Radio Frequency based systems for indoor positioning such as Radio Frequency Identification (RFID) is a key enabler technology for the IoT due to its costeffective, high readability rates, automatic identification and, importantly, its energy efficiency characteristic.This paper reviews the state-of-the-art RFID technologies in IoT Smart Homes applications.It presents several comparable studies of RFID based projects in smart ho …

Card 2 · Analysis of Robust Consensus Algorithms for $q$-Triangular $r$-Regular Networks

Internet of Things(IoT) is a heterogeneous network consists of various physical objects such as large number of sensors, actuators, RFID tags, smart devices, and servers connected to the internet. IoT networks have potential applications in healthcare, transportation, smart home, and automotive indu…

Read full passage excerpt

Internet of Things(IoT) is a heterogeneous network consists of various physical objects such as large number of sensors, actuators, RFID tags, smart devices, and servers connected to the internet. IoT networks have potential applications in healthcare, transportation, smart home, and automotive industries. To realize the IoT applications, all these devices need to be dynamically cooperated and utilize their resources effectively in a distributed fashion. Consensus algorithms have attracted much research attention in recent years due to their simple execution, robustness to topology changes, and distributed philosophy. These algorithms are extensively utilized for synchronization, resource allocation, and security in IoT networks. Performance of the distributed consensus algorithms can be effectively quantified by the Convergence Time, Network Coherence, Maximum Communication Time-Delay. In this work, we model the IoT network as a q-triangular r-regular ring network as q-triangular topo …

Card 3 · Health Monitoring in Smart Homes Utilizing Internet of Things

In recent years the concept of the Internet of Things (IoT) has evolved to connect commercial gadgets together with the medical field to facilitate an unprecedented range of accessibility. The development of medical devices connected to internet of things has been praised for the potential of allevi…

Read full passage excerpt

In recent years the concept of the Internet of Things (IoT) has evolved to connect commercial gadgets together with the medical field to facilitate an unprecedented range of accessibility. The development of medical devices connected to internet of things has been praised for the potential of alleviating the strain on the modern healthcare system by giving users the opportunity to reside in the home during treatment or recovery. With the IoT becoming more prevalent and available at a commercial level, there exists room for integration into emerging, intelligent environments such as smart homes. When used in tandem with conventional healthcare, the IoT offers a vast range of custom-tailored treatment options. This paper studies recent state-of-the-art research on the field of IoT for health monitoring and smart homes, examines several potential use-cases of blending the technology, and proposes integration with an existing smart home testbed for further study. Challenges of adoption and …

Card 4 · Smart Home, security concerns of IoT

The IoT (Internet of Things) has become widely popular in the domestic environments. People are renewing their homes into smart homes; however, the privacy concerns of owning many Internet connected devices with always-on environmental sensors remain insufficiently addressed. Default and weak passwo…

Read full passage excerpt

The IoT (Internet of Things) has become widely popular in the domestic environments. People are renewing their homes into smart homes; however, the privacy concerns of owning many Internet connected devices with always-on environmental sensors remain insufficiently addressed. Default and weak passwords, cheap materials and hardware, and unencrypted communication are identified as the principal threats and vulnerabilities of IoT devices. Solutions and countermeasures are also provided: choosing a strong password, strong authentication mechanisms, check online databases of exposed or default credentials to mitigate the first threat; a selection of smart home devices from reputable companies and the implementation of the SDN for the Dos/DDoS threat; and finally IDS, HTTPS protocol and VPN for eavesdropping. The paper concludes dealing with a further challenge, "the lack of technical support", by which an auto-configuration approach should be analysed; this could both ease the installation …

Card 5 · A Study on Internet of Things based Applications

This paper gives a detail analysis of various applications based on Internet of Thing (IoT)s. This explains about how internet of things evolved from mobile computing and ubiquitous computing. It emphasises the fact that objects are connected over the internet rather than people. The properties of I…

Read full passage excerpt

This paper gives a detail analysis of various applications based on Internet of Thing (IoT)s. This explains about how internet of things evolved from mobile computing and ubiquitous computing. It emphasises the fact that objects are connected over the internet rather than people. The properties of Internet of Things (IOT) are product information, electronic tag, standard expressed and uploading information. It utilises the Radio Frequency Identification (RFID) technology and wireless sensor networks (WSN). IOT applications are used in domains such as healthcare, supply chain management, defence and agriculture. Lastly the paper focuses on issues involved in IOT. Though it is a boon, IOT faces certain crucial issues like privacy and security.

Model-reported card use

Card 5 · fact

Affected output: domains such as healthcare, transportation, smart homes, and agriculture

Model explanation: Supplied the concrete list of IoT application domains used to specify where the proliferation of connected devices has extended.

Card 1 · terminology

Affected output: smart homes

Model explanation: Reinforced the smart-home framing already present in the draft's preceding sentence about connected household devices.

No Evidence

Same model and position · no retrieved passages · first output

Not preferred in this comparison

The Internet of Behavior (IoB) extends this connectivity by focusing on the data generated through human interaction with connected devices, aiming to understand and influence behavior rather than merely to connect objects.

Earlier card-use comparison

The earlier diagnostic compared only cases with reported card use and text on both sides; its scope differs from the overall AI review above.

The blinded model review preferred the no-Evidence version. Its original rationale appears below.

Read the original model judgment

Output A adds an unsupported claim that IoT data streams 'increasingly capture...human activity and behaviour' as a stated fact, which is not supported by any supplied source. The sources discuss IoT applications in various domains but do not state that data streams capture human activity and behavior, nor do they describe a trend toward this. Output B, by contrast, introduces the Internet of Behavior (IoB) as a conceptual extension of IoT connectivity, which is directly relevant to the manuscript's stated focus (a survey on IoB) and does not make unsupported factual claims about trends or data capture. The draft explicitly states this is 'the first survey paper on the Internet of Behavior,' making a transition to IoB natural and contextually grounded. B's claim about IoB's focus on 'data generated through human interaction with connected devices' is a definitional/conceptual framing that aligns with the paper's scope without requiring source support for a factual trend.

The passages are input material; their presence does not mean an output used them or that they support every claim. Draft and source passages are excerpted; no simulated scores or PDF appear here.